[ { "URL": "https://arxiv.org/abs/1602.02261v2", "archive": "arXiv", "archive_location": "1602.02261", "author": [ { "literal": "Rodrigo Nogueira" }, { "literal": "Kyunghyun Cho" } ], "citation-key": "arxiv_1602_02261", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "67f8697c21dabc2c4781eba46bfb60671506b413471efcdb28864c11649b3a2c", "submitted_at": "2016-02-06T14:53:02Z", "updated_at": "2016-05-20T16:26:58Z", "version": 2, "versioned_url": "https://arxiv.org/abs/1602.02261v2" }, "canonical_source_url": "https://arxiv.org/abs/1602.02261", "catalog_id": "arxiv:1602.02261", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2016-02-06", "identifiers": { "arxiv": "1602.02261" }, "record_updated_at": "2016-05-20" }, "genre": "preprint", "id": "arxiv:1602.02261", "issued": { "date-parts": [ [ 2016, 2, 6 ] ] }, "note": "First published: 2016-02-06; Recorded arXiv version: v2; Recorded version updated: 2016-05-20T16:26:58Z.", "title": "End-to-End Goal-Driven Web Navigation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/1808.09132v2", "archive": "arXiv", "archive_location": "1808.09132", "author": [ { "literal": "Panupong Pasupat" }, { "literal": "Tian-Shun Jiang" }, { "literal": "Evan Zheran Liu" }, { "literal": "Kelvin Guu" }, { "literal": "Percy Liang" } ], "citation-key": "arxiv_1808_09132", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f368116f70e987cde111835a05b5b36e0d345dabb4e8351610e2bbfb691b3440", "submitted_at": "2018-08-28T06:09:39Z", "updated_at": "2018-10-01T03:22:58Z", "version": 2, "versioned_url": "https://arxiv.org/abs/1808.09132v2" }, "canonical_source_url": "https://arxiv.org/abs/1808.09132", "catalog_id": "arxiv:1808.09132", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2018-08-28", "identifiers": { "arxiv": "1808.09132" }, "record_updated_at": "2018-10-01" }, "genre": "preprint", "id": "arxiv:1808.09132", "issued": { "date-parts": [ [ 2018, 8, 28 ] ] }, "note": "First published: 2018-08-28; Recorded arXiv version: v2; Recorded version updated: 2018-10-01T03:22:58Z.", "title": "Mapping Natural Language Commands to Web Elements", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2008.05132v2", "archive": "arXiv", "archive_location": "2008.05132", "author": [ { "literal": "Jieshan Chen" }, { "literal": "Mulong Xie" }, { "literal": "Zhenchang Xing" }, { "literal": "Chunyang Chen" }, { "literal": "Xiwei Xu" }, { "literal": "Liming Zhu" }, { "literal": "Guoqiang Li" } ], "citation-key": "arxiv_2008_05132", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d6352985532945fd562ce0382aef0e977ce1ec5909b6872e5fe048167e9732a0", "submitted_at": "2020-08-12T06:36:33Z", "updated_at": "2020-09-07T12:57:33Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2008.05132v2" }, "canonical_source_url": "https://arxiv.org/abs/2008.05132", "catalog_id": "arxiv:2008.05132", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2020-08-12", "identifiers": { "arxiv": "2008.05132" }, "record_updated_at": "2020-09-07" }, "genre": "preprint", "id": "arxiv:2008.05132", "issued": { "date-parts": [ [ 2020, 8, 12 ] ] }, "note": "First published: 2020-08-12; Recorded arXiv version: v2; Recorded version updated: 2020-09-07T12:57:33Z.", "title": "Object Detection for Graphical User Interface: Old Fashioned or Deep Learning or a Combination?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2101.09465v2", "archive": "arXiv", "archive_location": "2101.09465", "author": [ { "literal": "Xingyu Chen" }, { "literal": "Zihan Zhao" }, { "literal": "Lu Chen" }, { "literal": "Danyang Zhang" }, { "literal": "Jiabao Ji" }, { "literal": "Ao Luo" }, { "literal": "Yuxuan Xiong" }, { "literal": "Kai Yu" } ], "citation-key": "arxiv_2101_09465", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5fdf4bc31edd633fd0858835db0afedeef5dbc2694c4511903976c338bd19b1b", "submitted_at": "2021-01-23T09:43:44Z", "updated_at": "2021-11-08T08:31:44Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2101.09465v2" }, "canonical_source_url": "https://arxiv.org/abs/2101.09465", "catalog_id": "arxiv:2101.09465", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2021-01-23", "identifiers": { "arxiv": "2101.09465" }, "record_updated_at": "2021-11-08" }, "genre": "preprint", "id": "arxiv:2101.09465", "issued": { "date-parts": [ [ 2021, 1, 23 ] ] }, "note": "First published: 2021-01-23; Recorded arXiv version: v2; Recorded version updated: 2021-11-08T08:31:44Z.", "title": "WebSRC: A Dataset for Web-Based Structural Reading Comprehension", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2202.02312v3", "archive": "arXiv", "archive_location": "2202.02312", "author": [ { "literal": "Andrea Burns" }, { "literal": "Deniz Arsan" }, { "literal": "Sanjna Agrawal" }, { "literal": "Ranjitha Kumar" }, { "literal": "Kate Saenko" }, { "literal": "Bryan A. Plummer" } ], "citation-key": "arxiv_2202_02312", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1687369e31b0ad3632381d4a46b1d555c2e2ed1d6504e9c1b63006dd29f3445e", "submitted_at": "2022-02-04T18:51:50Z", "updated_at": "2022-08-15T00:24:24Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2202.02312v3" }, "canonical_source_url": "https://arxiv.org/abs/2202.02312", "catalog_id": "arxiv:2202.02312", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2022-02-04", "identifiers": { "arxiv": "2202.02312" }, "record_updated_at": "2022-08-15" }, "genre": "preprint", "id": "arxiv:2202.02312", "issued": { "date-parts": [ [ 2022, 2, 4 ] ] }, "note": "First published: 2022-02-04; Recorded arXiv version: v3; Recorded version updated: 2022-08-15T00:24:24Z.", "title": "A Dataset for Interactive Vision-Language Navigation with Unknown Command Feasibility", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2207.01206v4", "archive": "arXiv", "archive_location": "2207.01206", "author": [ { "literal": "Shunyu Yao" }, { "literal": "Howard Chen" }, { "literal": "John Yang" }, { "literal": "Karthik Narasimhan" } ], "citation-key": "arxiv_2207_01206", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6de4ab2a89875f1ccd99bbaa075c234d82979e845f27fe9530bdad0539c49570", "submitted_at": "2022-07-04T05:30:22Z", "updated_at": "2023-02-08T01:39:30Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2207.01206v4" }, "canonical_source_url": "https://arxiv.org/abs/2207.01206", "catalog_id": "arxiv:2207.01206", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2022-07-04", "identifiers": { "arxiv": "2207.01206" }, "record_updated_at": "2023-02-08" }, "genre": "preprint", "id": "arxiv:2207.01206", "issued": { "date-parts": [ [ 2022, 7, 4 ] ] }, "note": "First published: 2022-07-04; Recorded arXiv version: v4; Recorded version updated: 2023-02-08T01:39:30Z.", "title": "WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2209.08199v4", "archive": "arXiv", "archive_location": "2209.08199", "author": [ { "literal": "Yu-Chung Hsiao" }, { "literal": "Fedir Zubach" }, { "literal": "Gilles Baechler" }, { "literal": "Srinivas Sunkara" }, { "literal": "Victor Carbune" }, { "literal": "Jason Lin" }, { "literal": "Maria Wang" }, { "literal": "Yun Zhu" }, { "literal": "Jindong Chen" } ], "citation-key": "arxiv_2209_08199", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "37f7ba5d5b9bbc96fd3a723e49190f296e08d2874dcd545272d3fcef3cc33d24", "submitted_at": "2022-09-16T23:49:00Z", "updated_at": "2025-02-09T21:09:17Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2209.08199v4" }, "canonical_source_url": "https://arxiv.org/abs/2209.08199", "catalog_id": "arxiv:2209.08199", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2022-09-16", "identifiers": { "arxiv": "2209.08199" }, "record_updated_at": "2025-02-09" }, "genre": "preprint", "id": "arxiv:2209.08199", "issued": { "date-parts": [ [ 2022, 9, 16 ] ] }, "note": "First published: 2022-09-16; Recorded arXiv version: v4; Recorded version updated: 2025-02-09T21:09:17Z.", "title": "ScreenQA: Large-Scale Question-Answer Pairs over Mobile App Screenshots", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2305.08144v4", "archive": "arXiv", "archive_location": "2305.08144", "author": [ { "literal": "Danyang Zhang" }, { "literal": "Zhennan Shen" }, { "literal": "Rui Xie" }, { "literal": "Situo Zhang" }, { "literal": "Tianbao Xie" }, { "literal": "Zihan Zhao" }, { "literal": "Siyuan Chen" }, { "literal": "Lu Chen" }, { "literal": "Hongshen Xu" }, { "literal": "Ruisheng Cao" }, { "literal": "Kai Yu" } ], "citation-key": "arxiv_2305_08144", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8e454f547bdf8e01f47eb63c20f0cacfa8bd17a8c4ad37dfd3a1cceb732de178", "submitted_at": "2023-05-14T12:31:03Z", "updated_at": "2024-06-13T11:51:37Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2305.08144v4" }, "canonical_source_url": "https://arxiv.org/abs/2305.08144", "catalog_id": "arxiv:2305.08144", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-05-14", "identifiers": { "arxiv": "2305.08144" }, "record_updated_at": "2024-06-13" }, "genre": "preprint", "id": "arxiv:2305.08144", "issued": { "date-parts": [ [ 2023, 5, 14 ] ] }, "note": "First published: 2023-05-14; Recorded arXiv version: v4; Recorded version updated: 2024-06-13T11:51:37Z.", "title": "Mobile-Env: Building Qualified Evaluation Benchmarks for LLM-GUI Interaction", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2305.16653v1", "archive": "arXiv", "archive_location": "2305.16653", "author": [ { "literal": "Haotian Sun" }, { "literal": "Yuchen Zhuang" }, { "literal": "Lingkai Kong" }, { "literal": "Bo Dai" }, { "literal": "Chao Zhang" } ], "citation-key": "arxiv_2305_16653", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c254a26ebb8e1bcdca70b7581c780aa6b0dd4e5f5ded85e456f7e8fb8b38fcc6", "submitted_at": "2023-05-26T05:52:27Z", "updated_at": "2023-05-26T05:52:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2305.16653v1" }, "canonical_source_url": "https://arxiv.org/abs/2305.16653", "catalog_id": "arxiv:2305.16653", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2023-05-26", "identifiers": { "arxiv": "2305.16653" }, "record_updated_at": "2023-05-26" }, "genre": "preprint", "id": "arxiv:2305.16653", "issued": { "date-parts": [ [ 2023, 5, 26 ] ] }, "note": "First published: 2023-05-26; Recorded arXiv version: v1; Recorded version updated: 2023-05-26T05:52:27Z.", "title": "AdaPlanner: Adaptive Planning from Feedback with Language Models", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2305.19308v2", "archive": "arXiv", "archive_location": "2305.19308", "author": [ { "literal": "Hongxin Li" }, { "literal": "Jingran Su" }, { "literal": "Yuntao Chen" }, { "literal": "Qing Li" }, { "literal": "Zhaoxiang Zhang" } ], "citation-key": "arxiv_2305_19308", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "320ff40872677e9d32efae360dc289ca345076e037e7044188868f765e816008", "submitted_at": "2023-05-30T17:59:30Z", "updated_at": "2023-10-30T06:36:24Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2305.19308v2" }, "canonical_source_url": "https://arxiv.org/abs/2305.19308", "catalog_id": "arxiv:2305.19308", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-05-30", "identifiers": { "arxiv": "2305.19308" }, "record_updated_at": "2023-10-30" }, "genre": "preprint", "id": "arxiv:2305.19308", "issued": { "date-parts": [ [ 2023, 5, 30 ] ] }, "note": "First published: 2023-05-30; Recorded arXiv version: v2; Recorded version updated: 2023-10-30T06:36:24Z.", "title": "SheetCopilot: Bringing Software Productivity to the Next Level through Large Language Models", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2306.01242v2", "archive": "arXiv", "archive_location": "2306.01242", "author": [ { "literal": "Zhizheng Zhang" }, { "literal": "Xiaoyi Zhang" }, { "literal": "Wenxuan Xie" }, { "literal": "Yan Lu" } ], "citation-key": "arxiv_2306_01242", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "631a39d2929c68bb93a7f95af22c17d2ab208668fd97d9082fe0601a73d123dd", "submitted_at": "2023-06-02T02:42:58Z", "updated_at": "2023-12-04T13:36:59Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2306.01242v2" }, "canonical_source_url": "https://arxiv.org/abs/2306.01242", "catalog_id": "arxiv:2306.01242", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2023-06-02", "identifiers": { "arxiv": "2306.01242" }, "record_updated_at": "2023-12-04" }, "genre": "preprint", "id": "arxiv:2306.01242", "issued": { "date-parts": [ [ 2023, 6, 2 ] ] }, "note": "First published: 2023-06-02; Recorded arXiv version: v2; Recorded version updated: 2023-12-04T13:36:59Z.", "title": "Responsible Task Automation: Empowering Large Language Models as Responsible Task Automators", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2306.06070v3", "archive": "arXiv", "archive_location": "2306.06070", "author": [ { "literal": "Xiang Deng" }, { "literal": "Yu Gu" }, { "literal": "Boyuan Zheng" }, { "literal": "Shijie Chen" }, { "literal": "Samuel Stevens" }, { "literal": "Boshi Wang" }, { "literal": "Huan Sun" }, { "literal": "Yu Su" } ], "citation-key": "arxiv_2306_06070", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "de055dfa452130f3576839c99c007053c62a0c96c9bfbcdd79ecad1a9f74f830", "submitted_at": "2023-06-09T17:44:31Z", "updated_at": "2023-12-09T05:57:46Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2306.06070v3" }, "canonical_source_url": "https://arxiv.org/abs/2306.06070", "catalog_id": "arxiv:2306.06070", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-06-09", "identifiers": { "arxiv": "2306.06070" }, "record_updated_at": "2023-12-09" }, "genre": "preprint", "id": "arxiv:2306.06070", "issued": { "date-parts": [ [ 2023, 6, 9 ] ] }, "note": "First published: 2023-06-09; Recorded arXiv version: v3; Recorded version updated: 2023-12-09T05:57:46Z.", "title": "Mind2Web: Towards a Generalist Agent for the Web", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2307.10088v2", "archive": "arXiv", "archive_location": "2307.10088", "author": [ { "literal": "Christopher Rawles" }, { "literal": "Alice Li" }, { "literal": "Daniel Rodriguez" }, { "literal": "Oriana Riva" }, { "literal": "Timothy Lillicrap" } ], "citation-key": "arxiv_2307_10088", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "def4616f7cb13fe839e500c4b3171eafbe37b155b5d14301137806d6a5cde939", "submitted_at": "2023-07-19T15:57:24Z", "updated_at": "2023-10-27T14:24:31Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2307.10088v2" }, "canonical_source_url": "https://arxiv.org/abs/2307.10088", "catalog_id": "arxiv:2307.10088", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2023-07-19", "identifiers": { "arxiv": "2307.10088" }, "publication": { "authors": [ "Christopher Rawles", "Alice Li", "Daniel Rodriguez", "Oriana Riva", "Timothy Lillicrap" ], "date_basis": "Explicit publisher citation_publication_date: 2023-12-15. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2023, 12, 15 ], "doi": "10.52202/075280-2609", "event_dates": null, "evidence_urls": [ "https://proceedings.neurips.cc/paper_files/paper/2023/hash/bbbb6308b402fe909c39dd29950c32e0-Abstract-Datasets_and_Benchmarks.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "59708-59728", "pdf_url": "https://proceedings.neurips.cc/paper_files/paper/2023/file/bbbb6308b402fe909c39dd29950c32e0-Paper-Datasets_and_Benchmarks.pdf", "publisher": null, "series": null, "title": "AndroidInTheWild: A Large-Scale Dataset For Android Device Control", "type": "conference-paper", "url": "https://proceedings.neurips.cc/paper_files/paper/2023/hash/bbbb6308b402fe909c39dd29950c32e0-Abstract-Datasets_and_Benchmarks.html", "venue": "Advances in Neural Information Processing Systems", "volume": "36" }, "record_updated_at": "2023-10-27" }, "genre": "preprint", "id": "arxiv:2307.10088", "issued": { "date-parts": [ [ 2023, 7, 19 ] ] }, "note": "First published: 2023-07-19; Recorded arXiv version: v2; Recorded version updated: 2023-10-27T14:24:31Z.", "title": "Android in the Wild: A Large-Scale Dataset for Android Device Control", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2307.13854v4", "archive": "arXiv", "archive_location": "2307.13854", "author": [ { "literal": "Shuyan Zhou" }, { "literal": "Frank F. Xu" }, { "literal": "Hao Zhu" }, { "literal": "Xuhui Zhou" }, { "literal": "Robert Lo" }, { "literal": "Abishek Sridhar" }, { "literal": "Xianyi Cheng" }, { "literal": "Tianyue Ou" }, { "literal": "Yonatan Bisk" }, { "literal": "Daniel Fried" }, { "literal": "Uri Alon" }, { "literal": "Graham Neubig" } ], "citation-key": "arxiv_2307_13854", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0e9eca0165573ba53b663acbb1c7962fd2d64bc6633f10397f9cad229772a363", "submitted_at": "2023-07-25T22:59:32Z", "updated_at": "2024-04-16T15:13:18Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2307.13854v4" }, "canonical_source_url": "https://arxiv.org/abs/2307.13854", "catalog_id": "arxiv:2307.13854", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-07-25", "identifiers": { "arxiv": "2307.13854" }, "record_updated_at": "2024-04-16" }, "genre": "preprint", "id": "arxiv:2307.13854", "issued": { "date-parts": [ [ 2023, 7, 25 ] ] }, "note": "First published: 2023-07-25; Recorded arXiv version: v4; Recorded version updated: 2024-04-16T15:13:18Z.", "title": "WebArena: A Realistic Web Environment for Building Autonomous Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2308.03688v3", "archive": "arXiv", "archive_location": "2308.03688", "author": [ { "literal": "Xiao Liu" }, { "literal": "Hao Yu" }, { "literal": "Hanchen Zhang" }, { "literal": "Yifan Xu" }, { "literal": "Xuanyu Lei" }, { "literal": "Hanyu Lai" }, { "literal": "Yu Gu" }, { "literal": "Hangliang Ding" }, { "literal": "Kaiwen Men" }, { "literal": "Kejuan Yang" }, { "literal": "Shudan Zhang" }, { "literal": "Xiang Deng" }, { "literal": "Aohan Zeng" }, { "literal": "Zhengxiao Du" }, { "literal": "Chenhui Zhang" }, { "literal": "Sheng Shen" }, { "literal": "Tianjun Zhang" }, { "literal": "Yu Su" }, { "literal": "Huan Sun" }, { "literal": "Minlie Huang" }, { "literal": "Yuxiao Dong" }, { "literal": "Jie Tang" } ], "citation-key": "arxiv_2308_03688", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d8c4ba4db4aeed3a7b7ce7f77d610d7e34b0e8db26a36463e2c0bdffcad8f22a", "submitted_at": "2023-08-07T16:08:11Z", "updated_at": "2025-10-04T03:54:18Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2308.03688v3" }, "canonical_source_url": "https://arxiv.org/abs/2308.03688", "catalog_id": "arxiv:2308.03688", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2023-08-07", "identifiers": { "arxiv": "2308.03688" }, "record_updated_at": "2025-10-04" }, "genre": "preprint", "id": "arxiv:2308.03688", "issued": { "date-parts": [ [ 2023, 8, 7 ] ] }, "note": "First published: 2023-08-07; Recorded arXiv version: v3; Recorded version updated: 2025-10-04T03:54:18Z.", "title": "AgentBench: Evaluating LLMs as Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2310.14495v1", "archive": "arXiv", "archive_location": "2310.14495", "author": [ { "literal": "Justin Payan" }, { "literal": "Swaroop Mishra" }, { "literal": "Mukul Singh" }, { "literal": "Carina Negreanu" }, { "literal": "Christian Poelitz" }, { "literal": "Chitta Baral" }, { "literal": "Subhro Roy" }, { "literal": "Rasika Chakravarthy" }, { "literal": "Benjamin Van Durme" }, { "literal": "Elnaz Nouri" } ], "citation-key": "arxiv_2310_14495", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1648acae2228b8ae2f042d7986585a66bc6a649320d569b12fd093b96c15cb5c", "submitted_at": "2023-10-23T02:00:55Z", "updated_at": "2023-10-23T02:00:55Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2310.14495v1" }, "canonical_source_url": "https://arxiv.org/abs/2310.14495", "catalog_id": "arxiv:2310.14495", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2023-10-23", "identifiers": { "arxiv": "2310.14495" }, "record_updated_at": "2023-10-23" }, "genre": "preprint", "id": "arxiv:2310.14495", "issued": { "date-parts": [ [ 2023, 10, 23 ] ] }, "note": "First published: 2023-10-23; Recorded arXiv version: v1; Recorded version updated: 2023-10-23T02:00:55Z.", "title": "InstructExcel: A Benchmark for Natural Language Instruction in Excel", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2310.18331v2", "archive": "arXiv", "archive_location": "2310.18331", "author": [ { "literal": "Jiarun Liu" }, { "literal": "Wentao Hu" }, { "literal": "Chunhong Zhang" } ], "citation-key": "arxiv_2310_18331", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c72eab40cdc28981cf83a753d88f5181b5441d82cd568265457fef96d75b36b6", "submitted_at": "2023-10-20T11:10:14Z", "updated_at": "2023-10-31T06:25:59Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2310.18331v2" }, "canonical_source_url": "https://arxiv.org/abs/2310.18331", "catalog_id": "arxiv:2310.18331", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-10-20", "identifiers": { "arxiv": "2310.18331" }, "record_updated_at": "2023-10-31" }, "genre": "preprint", "id": "arxiv:2310.18331", "issued": { "date-parts": [ [ 2023, 10, 20 ] ] }, "note": "First published: 2023-10-20; Recorded arXiv version: v2; Recorded version updated: 2023-10-31T06:25:59Z.", "title": "AllTogether: Investigating the Efficacy of Spliced Prompt for Web Navigation using Large Language Models", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2311.01767v2", "archive": "arXiv", "archive_location": "2311.01767", "author": [ { "literal": "Yiduo Guo" }, { "literal": "Zekai Zhang" }, { "literal": "Yaobo Liang" }, { "literal": "Dongyan Zhao" }, { "literal": "Nan Duan" } ], "citation-key": "arxiv_2311_01767", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "58dfa75fd82a2cc1d41954f10feb1d987de6469782d11d04c3b8bfffc8a3b745", "submitted_at": "2023-11-03T08:06:35Z", "updated_at": "2023-11-07T10:13:34Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2311.01767v2" }, "canonical_source_url": "https://arxiv.org/abs/2311.01767", "catalog_id": "arxiv:2311.01767", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2023-11-03", "identifiers": { "arxiv": "2311.01767" }, "record_updated_at": "2023-11-07" }, "genre": "preprint", "id": "arxiv:2311.01767", "issued": { "date-parts": [ [ 2023, 11, 3 ] ] }, "note": "First published: 2023-11-03; Recorded arXiv version: v2; Recorded version updated: 2023-11-07T10:13:34Z.", "title": "PPTC Benchmark: Evaluating Large Language Models for PowerPoint Task Completion", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2311.12983v1", "archive": "arXiv", "archive_location": "2311.12983", "author": [ { "literal": "Grégoire Mialon" }, { "literal": "Clémentine Fourrier" }, { "literal": "Craig Swift" }, { "literal": "Thomas Wolf" }, { "literal": "Yann LeCun" }, { "literal": "Thomas Scialom" } ], "citation-key": "arxiv_2311_12983", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "29ddf2b7b4d495ee18bbd263cd77f86a71b3014b58a6f140c3bfce7cbae9c922", "submitted_at": "2023-11-21T20:34:47Z", "updated_at": "2023-11-21T20:34:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2311.12983v1" }, "canonical_source_url": "https://arxiv.org/abs/2311.12983", "catalog_id": "arxiv:2311.12983", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2023-11-21", "identifiers": { "arxiv": "2311.12983" }, "record_updated_at": "2023-11-21" }, "genre": "preprint", "id": "arxiv:2311.12983", "issued": { "date-parts": [ [ 2023, 11, 21 ] ] }, "note": "First published: 2023-11-21; Recorded arXiv version: v1; Recorded version updated: 2023-11-21T20:34:47Z.", "title": "GAIA: a benchmark for General AI Assistants", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2311.18751v3", "archive": "arXiv", "archive_location": "2311.18751", "author": [ { "literal": "Hiroki Furuta" }, { "literal": "Yutaka Matsuo" }, { "literal": "Aleksandra Faust" }, { "literal": "Izzeddin Gur" } ], "citation-key": "arxiv_2311_18751", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4b00e0a9c74b7ac8819bdda540a1069c33cbac45f0c7fb853dee995f2feb9a59", "submitted_at": "2023-11-30T17:50:47Z", "updated_at": "2024-12-31T04:24:16Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2311.18751v3" }, "canonical_source_url": "https://arxiv.org/abs/2311.18751", "catalog_id": "arxiv:2311.18751", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-11-30", "identifiers": { "arxiv": "2311.18751" }, "record_updated_at": "2024-12-31" }, "genre": "preprint", "id": "arxiv:2311.18751", "issued": { "date-parts": [ [ 2023, 11, 30 ] ] }, "note": "First published: 2023-11-30; Recorded arXiv version: v3; Recorded version updated: 2024-12-31T04:24:16Z.", "title": "Exposing Limitations of Language Model Agents in Sequential-Task Compositions on the Web", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2312.13108v2", "archive": "arXiv", "archive_location": "2312.13108", "author": [ { "literal": "Difei Gao" }, { "literal": "Lei Ji" }, { "literal": "Zechen Bai" }, { "literal": "Mingyu Ouyang" }, { "literal": "Peiran Li" }, { "literal": "Dongxing Mao" }, { "literal": "Qinchen Wu" }, { "literal": "Weichen Zhang" }, { "literal": "Peiyi Wang" }, { "literal": "Xiangwu Guo" }, { "literal": "Hengxu Wang" }, { "literal": "Luowei Zhou" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2312_13108", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c771b076044c170728041a4f54662d21493adcc3742d40db08cb2f9bd0614655", "submitted_at": "2023-12-20T15:28:38Z", "updated_at": "2024-01-01T14:26:39Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2312.13108v2" }, "canonical_source_url": "https://arxiv.org/abs/2312.13108", "catalog_id": "arxiv:2312.13108", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-12-20", "identifiers": { "arxiv": "2312.13108" }, "record_updated_at": "2024-01-01" }, "genre": "preprint", "id": "arxiv:2312.13108", "issued": { "date-parts": [ [ 2023, 12, 20 ] ] }, "note": "First published: 2023-12-20; Recorded arXiv version: v2; Recorded version updated: 2024-01-01T14:26:39Z.", "title": "ASSISTGUI: Task-Oriented Desktop Graphical User Interface Automation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2312.15820v1", "archive": "arXiv", "archive_location": "2312.15820", "author": [ { "literal": "Qi Chen" }, { "literal": "Dileepa Pitawela" }, { "literal": "Chongyang Zhao" }, { "literal": "Gengze Zhou" }, { "literal": "Hsiang-Ting Chen" }, { "literal": "Qi Wu" } ], "citation-key": "arxiv_2312_15820", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "47dedddced8f06bf436ab5a101d1ec4ecf4380c710e457dc7a35fc86d2e1f230", "submitted_at": "2023-12-25T22:13:26Z", "updated_at": "2023-12-25T22:13:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2312.15820v1" }, "canonical_source_url": "https://arxiv.org/abs/2312.15820", "catalog_id": "arxiv:2312.15820", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2023-12-25", "identifiers": { "arxiv": "2312.15820" }, "record_updated_at": "2023-12-25" }, "genre": "preprint", "id": "arxiv:2312.15820", "issued": { "date-parts": [ [ 2023, 12, 25 ] ] }, "note": "First published: 2023-12-25; Recorded arXiv version: v1; Recorded version updated: 2023-12-25T22:13:26Z.", "title": "WebVLN: Vision-and-Language Navigation on Websites", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2401.13178v2", "archive": "arXiv", "archive_location": "2401.13178", "author": [ { "literal": "Chang Ma" }, { "literal": "Junlei Zhang" }, { "literal": "Zhihao Zhu" }, { "literal": "Cheng Yang" }, { "literal": "Yujiu Yang" }, { "literal": "Yaohui Jin" }, { "literal": "Zhenzhong Lan" }, { "literal": "Lingpeng Kong" }, { "literal": "Junxian He" } ], "citation-key": "arxiv_2401_13178", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "22c5a60f9ed7a8488ec31fbf24d392e9674b0cae79ab1b269fd9d57202025801", "submitted_at": "2024-01-24T01:51:00Z", "updated_at": "2024-12-23T20:12:48Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2401.13178v2" }, "canonical_source_url": "https://arxiv.org/abs/2401.13178", "catalog_id": "arxiv:2401.13178", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-01-24", "identifiers": { "arxiv": "2401.13178" }, "record_updated_at": "2024-12-23" }, "genre": "preprint", "id": "arxiv:2401.13178", "issued": { "date-parts": [ [ 2024, 1, 24 ] ] }, "note": "First published: 2024-01-24; Recorded arXiv version: v2; Recorded version updated: 2024-12-23T20:12:48Z.", "title": "AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2401.13649v2", "archive": "arXiv", "archive_location": "2401.13649", "author": [ { "literal": "Jing Yu Koh" }, { "literal": "Robert Lo" }, { "literal": "Lawrence Jang" }, { "literal": "Vikram Duvvur" }, { "literal": "Ming Chong Lim" }, { "literal": "Po-Yu Huang" }, { "literal": "Graham Neubig" }, { "literal": "Shuyan Zhou" }, { "literal": "Ruslan Salakhutdinov" }, { "literal": "Daniel Fried" } ], "citation-key": "arxiv_2401_13649", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3a705b87b43464120ebb01b28b0f6a9e31fad68f3d463c88354f2005dab940c3", "submitted_at": "2024-01-24T18:35:21Z", "updated_at": "2024-06-06T02:01:09Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2401.13649v2" }, "canonical_source_url": "https://arxiv.org/abs/2401.13649", "catalog_id": "arxiv:2401.13649", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-01-24", "identifiers": { "arxiv": "2401.13649" }, "publication": { "authors": [ "Jing Yu Koh", "Robert Lo", "Lawrence Jang", "Vikram Duvvur", "Ming Chong Lim", "Po-Yu Huang", "Graham Neubig", "Shuyan Zhou", "Ruslan Salakhutdinov", "Daniel Fried" ], "date_basis": "Explicit publisher metadata: 2024/8; precision=month.", "date_parts": [ 2024, 8 ], "doi": "10.18653/v1/2024.acl-long.50", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2024.acl-long.50/", "https://aclanthology.org/2024.acl-long.50.pdf" ], "identity_basis": "Same work: exact normalized landing title, corresponding author list, and matching specific contribution. Publisher and arXiv versions remain distinct.", "issue": null, "notes": "The formal PDF title adds Visually Grounded; the landing metadata uses Visual Web Tasks. PDF authors include Ming Chong Lim and Ruslan Salakhutdinov; the landing abbreviates them to Ming Lim and Russ Salakhutdinov. Publication title/authors use the PDF; the retained arXiv title remains unchanged. Publisher-hosted proceedings record; distinct from the retained arXiv preprint snapshot.", "pages": "881-905", "pdf_url": "https://aclanthology.org/2024.acl-long.50.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "VisualWebArena: Evaluating Multimodal Agents on Realistic Visually Grounded Web Tasks", "type": "conference-paper", "url": "https://aclanthology.org/2024.acl-long.50/", "venue": "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", "volume": null }, "record_updated_at": "2024-06-06" }, "genre": "preprint", "id": "arxiv:2401.13649", "issued": { "date-parts": [ [ 2024, 1, 24 ] ] }, "note": "First published: 2024-01-24; Recorded arXiv version: v2; Recorded version updated: 2024-06-06T02:01:09Z.", "title": "VisualWebArena: Evaluating Multimodal Agents on Realistic Visual Web Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2401.13919v4", "archive": "arXiv", "archive_location": "2401.13919", "author": [ { "literal": "Hongliang He" }, { "literal": "Wenlin Yao" }, { "literal": "Kaixin Ma" }, { "literal": "Wenhao Yu" }, { "literal": "Yong Dai" }, { "literal": "Hongming Zhang" }, { "literal": "Zhenzhong Lan" }, { "literal": "Dong Yu" } ], "citation-key": "arxiv_2401_13919", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "dbc7121c9606c5987df0941b849c216ad8731bebcc74a061cf3f154379d7b8da", "submitted_at": "2024-01-25T03:33:18Z", "updated_at": "2024-06-06T18:37:34Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2401.13919v4" }, "canonical_source_url": "https://arxiv.org/abs/2401.13919", "catalog_id": "arxiv:2401.13919", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-01-25", "identifiers": { "arxiv": "2401.13919" }, "publication": { "authors": [ "Hongliang He", "Wenlin Yao", "Kaixin Ma", "Wenhao Yu", "Yong Dai", "Hongming Zhang", "Zhenzhong Lan", "Dong Yu" ], "date_basis": "Explicit publisher citation_publication_date: 2024/8. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 8 ], "doi": "10.18653/v1/2024.acl-long.371", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2024.acl-long.371/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity. The normalized publisher abstract matches the retained arXiv abstract SHA-256.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "6864-6890", "pdf_url": "https://aclanthology.org/2024.acl-long.371.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models", "type": "conference-paper", "url": "https://aclanthology.org/2024.acl-long.371/", "venue": "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", "volume": null }, "record_updated_at": "2024-06-06" }, "genre": "preprint", "id": "arxiv:2401.13919", "issued": { "date-parts": [ [ 2024, 1, 25 ] ] }, "note": "First published: 2024-01-25; Recorded arXiv version: v4; Recorded version updated: 2024-06-06T18:37:34Z.", "title": "WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2402.05930v2", "archive": "arXiv", "archive_location": "2402.05930", "author": [ { "literal": "Xing Han Lù" }, { "literal": "Zdeněk Kasner" }, { "literal": "Siva Reddy" } ], "citation-key": "arxiv_2402_05930", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6a7c1a1f260f9287e8999525e1adc5534a70f7b96271d469973fceee1ac23609", "submitted_at": "2024-02-08T18:58:02Z", "updated_at": "2024-09-10T22:22:43Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2402.05930v2" }, "canonical_source_url": "https://arxiv.org/abs/2402.05930", "catalog_id": "arxiv:2402.05930", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-02-08", "identifiers": { "arxiv": "2402.05930" }, "publication": { "authors": [ "Xing Han Lù", "Zdeněk Kasner", "Siva Reddy" ], "date_basis": "Explicit publisher metadata: 2024/07/08; precision=day.", "date_parts": [ 2024, 7, 8 ], "doi": null, "event_dates": "21--27 Jul 2024", "evidence_urls": [ "https://proceedings.mlr.press/v235/lu24e.html", "https://raw.githubusercontent.com/mlresearch/v235/main/assets/lu24e/lu24e.pdf" ], "identity_basis": "Same work: exact normalized landing title, corresponding author list, and matching specific contribution. Publisher and arXiv versions remain distinct.", "issue": null, "notes": "The PMLR landing renders Xing Han Lu; the paper title page has Xing Han Lù. Publication authors use the PDF. The publisher issue date comes from citation_publication_date; the conference date range is retained separately and does not replace it. Publisher-hosted proceedings record; distinct from the retained arXiv preprint snapshot.", "pages": "33007-33056", "pdf_url": "https://raw.githubusercontent.com/mlresearch/v235/main/assets/lu24e/lu24e.pdf", "publisher": "PMLR", "series": "Proceedings of Machine Learning Research", "title": "WebLINX: Real-World Website Navigation with Multi-Turn Dialogue", "type": "conference-paper", "url": "https://proceedings.mlr.press/v235/lu24e.html", "venue": "Proceedings of the 41st International Conference on Machine Learning", "volume": "235" }, "record_updated_at": "2024-09-10" }, "genre": "preprint", "id": "arxiv:2402.05930", "issued": { "date-parts": [ [ 2024, 2, 8 ] ] }, "note": "First published: 2024-02-08; Recorded arXiv version: v2; Recorded version updated: 2024-09-10T22:22:43Z.", "title": "WebLINX: Real-World Website Navigation with Multi-Turn Dialogue", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2402.06596v1", "archive": "arXiv", "archive_location": "2402.06596", "author": [ { "literal": "Mingzhe Xing" }, { "literal": "Rongkai Zhang" }, { "literal": "Hui Xue" }, { "literal": "Qi Chen" }, { "literal": "Fan Yang" }, { "literal": "Zhen Xiao" } ], "citation-key": "arxiv_2402_06596", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0504cfffc5d46ecf447a4fdeac561aede9482bb8640fcfd9d7f02cf6eaf47597", "submitted_at": "2024-02-09T18:19:25Z", "updated_at": "2024-02-09T18:19:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2402.06596v1" }, "canonical_source_url": "https://arxiv.org/abs/2402.06596", "catalog_id": "arxiv:2402.06596", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-02-09", "identifiers": { "arxiv": "2402.06596" }, "record_updated_at": "2024-02-09" }, "genre": "preprint", "id": "arxiv:2402.06596", "issued": { "date-parts": [ [ 2024, 2, 9 ] ] }, "note": "First published: 2024-02-09; Recorded arXiv version: v1; Recorded version updated: 2024-02-09T18:19:25Z.", "title": "Understanding the Weakness of Large Language Model Agents within a Complex Android Environment", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2402.15057v1", "archive": "arXiv", "archive_location": "2402.15057", "author": [ { "literal": "Yang Deng" }, { "literal": "Xuan Zhang" }, { "literal": "Wenxuan Zhang" }, { "literal": "Yifei Yuan" }, { "literal": "See-Kiong Ng" }, { "literal": "Tat-Seng Chua" } ], "citation-key": "arxiv_2402_15057", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5e4b233e423e9043b6d615c7b185a1eb0d47f646590e4fb9fd43ca156a1a1b16", "submitted_at": "2024-02-23T02:18:12Z", "updated_at": "2024-02-23T02:18:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2402.15057v1" }, "canonical_source_url": "https://arxiv.org/abs/2402.15057", "catalog_id": "arxiv:2402.15057", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-02-23", "identifiers": { "arxiv": "2402.15057" }, "publication": { "authors": [ "Yang Deng", "Xuan Zhang", "Wenxuan Zhang", "Yifei Yuan", "See Kiong Ng", "Tat-Seng Chua" ], "date_basis": "Explicit publisher citation_publication_date: 2024/8. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 8 ], "doi": "10.18653/v1/2024.acl-long.477", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2024.acl-long.477/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity. The normalized publisher abstract matches the retained arXiv abstract SHA-256.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "8795-8812", "pdf_url": "https://aclanthology.org/2024.acl-long.477.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "On the Multi-turn Instruction Following for Conversational Web Agents", "type": "conference-paper", "url": "https://aclanthology.org/2024.acl-long.477/", "venue": "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", "volume": null }, "record_updated_at": "2024-02-23" }, "genre": "preprint", "id": "arxiv:2402.15057", "issued": { "date-parts": [ [ 2024, 2, 23 ] ] }, "note": "First published: 2024-02-23; Recorded arXiv version: v1; Recorded version updated: 2024-02-23T02:18:12Z.", "title": "On the Multi-turn Instruction Following for Conversational Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2402.16965v1", "archive": "arXiv", "archive_location": "2402.16965", "author": [ { "literal": "Fangzhou Wu" }, { "literal": "Shutong Wu" }, { "literal": "Yulong Cao" }, { "literal": "Chaowei Xiao" } ], "citation-key": "arxiv_2402_16965", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2cf1458c8c4134b2195394a8acba4919f9b3a5a7c79b1bea017ac4558123f69c", "submitted_at": "2024-02-26T19:01:54Z", "updated_at": "2024-02-26T19:01:54Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2402.16965v1" }, "canonical_source_url": "https://arxiv.org/abs/2402.16965", "catalog_id": "arxiv:2402.16965", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-02-26", "identifiers": { "arxiv": "2402.16965" }, "record_updated_at": "2024-02-26" }, "genre": "preprint", "id": "arxiv:2402.16965", "issued": { "date-parts": [ [ 2024, 2, 26 ] ] }, "note": "First published: 2024-02-26; Recorded arXiv version: v1; Recorded version updated: 2024-02-26T19:01:54Z.", "title": "WIPI: A New Web Threat for LLM-Driven Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2402.17553v3", "archive": "arXiv", "archive_location": "2402.17553", "author": [ { "literal": "Raghav Kapoor" }, { "literal": "Yash Parag Butala" }, { "literal": "Melisa Russak" }, { "literal": "Jing Yu Koh" }, { "literal": "Kiran Kamble" }, { "literal": "Waseem Alshikh" }, { "literal": "Ruslan Salakhutdinov" } ], "citation-key": "arxiv_2402_17553", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "54b548ac2bee529d3a769e7ea14b6bfce2270cdf349d832cc8c146e7d779b81a", "submitted_at": "2024-02-27T14:47:53Z", "updated_at": "2024-07-21T23:16:13Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2402.17553v3" }, "canonical_source_url": "https://arxiv.org/abs/2402.17553", "catalog_id": "arxiv:2402.17553", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-02-27", "identifiers": { "arxiv": "2402.17553" }, "record_updated_at": "2024-07-21" }, "genre": "preprint", "id": "arxiv:2402.17553", "issued": { "date-parts": [ [ 2024, 2, 27 ] ] }, "note": "First published: 2024-02-27; Recorded arXiv version: v3; Recorded version updated: 2024-07-21T23:16:13Z.", "title": "OmniACT: A Dataset and Benchmark for Enabling Multimodal Generalist Autonomous Agents for Desktop and Web", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2403.03788v1", "archive": "arXiv", "archive_location": "2403.03788", "author": [ { "literal": "Zekai Zhang" }, { "literal": "Yiduo Guo" }, { "literal": "Yaobo Liang" }, { "literal": "Dongyan Zhao" }, { "literal": "Nan Duan" } ], "citation-key": "arxiv_2403_03788", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "51abf6b165b84c647c9a4870a6441900cac5e9a933199fe8a47d1b7f065eb11e", "submitted_at": "2024-03-06T15:33:32Z", "updated_at": "2024-03-06T15:33:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2403.03788v1" }, "canonical_source_url": "https://arxiv.org/abs/2403.03788", "catalog_id": "arxiv:2403.03788", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-03-06", "identifiers": { "arxiv": "2403.03788" }, "record_updated_at": "2024-03-06" }, "genre": "preprint", "id": "arxiv:2403.03788", "issued": { "date-parts": [ [ 2024, 3, 6 ] ] }, "note": "First published: 2024-03-06; Recorded arXiv version: v1; Recorded version updated: 2024-03-06T15:33:32Z.", "title": "PPTC-R benchmark: Towards Evaluating the Robustness of Large Language Models for PowerPoint Task Completion", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2403.07718v5", "archive": "arXiv", "archive_location": "2403.07718", "author": [ { "literal": "Alexandre Drouin" }, { "literal": "Maxime Gasse" }, { "literal": "Massimo Caccia" }, { "literal": "Issam H. Laradji" }, { "literal": "Manuel Del Verme" }, { "literal": "Tom Marty" }, { "literal": "Léo Boisvert" }, { "literal": "Megh Thakkar" }, { "literal": "Quentin Cappart" }, { "literal": "David Vazquez" }, { "literal": "Nicolas Chapados" }, { "literal": "Alexandre Lacoste" } ], "citation-key": "arxiv_2403_07718", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8033f9c26a5f95a49000374a6edebb4dead83e100ebcf5226007f06b7573c305", "submitted_at": "2024-03-12T14:58:45Z", "updated_at": "2024-07-23T06:19:28Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2403.07718v5" }, "canonical_source_url": "https://arxiv.org/abs/2403.07718", "catalog_id": "arxiv:2403.07718", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-03-12", "identifiers": { "arxiv": "2403.07718" }, "record_updated_at": "2024-07-23" }, "genre": "preprint", "id": "arxiv:2403.07718", "issued": { "date-parts": [ [ 2024, 3, 12 ] ] }, "note": "First published: 2024-03-12; Recorded arXiv version: v5; Recorded version updated: 2024-07-23T06:19:28Z.", "title": "WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2403.11905v4", "archive": "arXiv", "archive_location": "2403.11905", "author": [ { "literal": "Kevin Xu" }, { "literal": "Yeganeh Kordi" }, { "literal": "Tanay Nayak" }, { "literal": "Adi Asija" }, { "literal": "Yizhong Wang" }, { "literal": "Kate Sanders" }, { "literal": "Adam Byerly" }, { "literal": "Jingyu Zhang" }, { "literal": "Benjamin Van Durme" }, { "literal": "Daniel Khashabi" } ], "citation-key": "arxiv_2403_11905", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d82424b84ebddfeb005ffae38c130c6a9c7dae7adcdc59e2b7d001284b785573", "submitted_at": "2024-03-18T16:06:30Z", "updated_at": "2025-02-22T01:33:54Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2403.11905v4" }, "canonical_source_url": "https://arxiv.org/abs/2403.11905", "catalog_id": "arxiv:2403.11905", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-03-18", "identifiers": { "arxiv": "2403.11905" }, "publication": { "authors": [ "Kevin Xu", "Yeganeh Kordi", "Tanay Nayak", "Adi Asija", "Yizhong Wang", "Kate Sanders", "Adam Byerly", "Jingyu Zhang", "Benjamin Van Durme", "Daniel Khashabi" ], "date_basis": "Explicit publisher citation_publication_date: 2025/4. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 4 ], "doi": "10.18653/v1/2025.naacl-long.188", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.naacl-long.188/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "3694-3710", "pdf_url": "https://aclanthology.org/2025.naacl-long.188.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "TurkingBench: A Challenge Benchmark for Web Agents", "type": "conference-paper", "url": "https://aclanthology.org/2025.naacl-long.188/", "venue": "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)", "volume": null }, "record_updated_at": "2025-02-22" }, "genre": "preprint", "id": "arxiv:2403.11905", "issued": { "date-parts": [ [ 2024, 3, 18 ] ] }, "note": "First published: 2024-03-18; Recorded arXiv version: v4; Recorded version updated: 2025-02-22T01:33:54Z.", "title": "Tur[k]ingBench: A Challenge Benchmark for Web Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2403.17918v3", "archive": "arXiv", "archive_location": "2403.17918", "author": [ { "literal": "Longtao Zheng" }, { "literal": "Zhiyuan Huang" }, { "literal": "Zhenghai Xue" }, { "literal": "Xinrun Wang" }, { "literal": "Bo An" }, { "literal": "Shuicheng Yan" } ], "citation-key": "arxiv_2403_17918", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fe1432b46c15566182418bf3fa2f918607e8b0e9a383459416d03e28dc7e347c", "submitted_at": "2024-03-26T17:54:15Z", "updated_at": "2025-02-14T08:13:39Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2403.17918v3" }, "canonical_source_url": "https://arxiv.org/abs/2403.17918", "catalog_id": "arxiv:2403.17918", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-03-26", "identifiers": { "arxiv": "2403.17918" }, "record_updated_at": "2025-02-14" }, "genre": "preprint", "id": "arxiv:2403.17918", "issued": { "date-parts": [ [ 2024, 3, 26 ] ] }, "note": "First published: 2024-03-26; Recorded arXiv version: v3; Recorded version updated: 2025-02-14T08:13:39Z.", "title": "AgentStudio: A Toolkit for Building General Virtual Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2404.05955v1", "archive": "arXiv", "archive_location": "2404.05955", "author": [ { "literal": "Junpeng Liu" }, { "literal": "Yifan Song" }, { "literal": "Bill Yuchen Lin" }, { "literal": "Wai Lam" }, { "literal": "Graham Neubig" }, { "literal": "Yuanzhi Li" }, { "literal": "Xiang Yue" } ], "citation-key": "arxiv_2404_05955", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7414bfa289baf86a8f2bb54f3f69f25a6a4cc8516fb0296a1899d21ad3020d4f", "submitted_at": "2024-04-09T02:29:39Z", "updated_at": "2024-04-09T02:29:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2404.05955v1" }, "canonical_source_url": "https://arxiv.org/abs/2404.05955", "catalog_id": "arxiv:2404.05955", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-09", "identifiers": { "arxiv": "2404.05955" }, "record_updated_at": "2024-04-09" }, "genre": "preprint", "id": "arxiv:2404.05955", "issued": { "date-parts": [ [ 2024, 4, 9 ] ] }, "note": "First published: 2024-04-09; Recorded arXiv version: v1; Recorded version updated: 2024-04-09T02:29:39Z.", "title": "VisualWebBench: How Far Have Multimodal LLMs Evolved in Web Page Understanding and Grounding?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2404.06474v3", "archive": "arXiv", "archive_location": "2404.06474", "author": [ { "literal": "Jiayi Pan" }, { "literal": "Yichi Zhang" }, { "literal": "Nicholas Tomlin" }, { "literal": "Yifei Zhou" }, { "literal": "Sergey Levine" }, { "literal": "Alane Suhr" } ], "citation-key": "arxiv_2404_06474", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "abc75ee55c7ac5992656dcec2106e286c670a58130a59b0eef7f5fd74d7496c5", "submitted_at": "2024-04-09T17:25:47Z", "updated_at": "2024-10-07T14:19:37Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2404.06474v3" }, "canonical_source_url": "https://arxiv.org/abs/2404.06474", "catalog_id": "arxiv:2404.06474", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-09", "identifiers": { "arxiv": "2404.06474" }, "record_updated_at": "2024-10-07" }, "genre": "preprint", "id": "arxiv:2404.06474", "issued": { "date-parts": [ [ 2024, 4, 9 ] ] }, "note": "First published: 2024-04-09; Recorded arXiv version: v3; Recorded version updated: 2024-10-07T14:19:37Z.", "title": "Autonomous Evaluation and Refinement of Digital Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2404.07972v2", "archive": "arXiv", "archive_location": "2404.07972", "author": [ { "literal": "Tianbao Xie" }, { "literal": "Danyang Zhang" }, { "literal": "Jixuan Chen" }, { "literal": "Xiaochuan Li" }, { "literal": "Siheng Zhao" }, { "literal": "Ruisheng Cao" }, { "literal": "Toh Jing Hua" }, { "literal": "Zhoujun Cheng" }, { "literal": "Dongchan Shin" }, { "literal": "Fangyu Lei" }, { "literal": "Yitao Liu" }, { "literal": "Yiheng Xu" }, { "literal": "Shuyan Zhou" }, { "literal": "Silvio Savarese" }, { "literal": "Caiming Xiong" }, { "literal": "Victor Zhong" }, { "literal": "Tao Yu" } ], "citation-key": "arxiv_2404_07972", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a347bfcb4de03a60410392b52d522e5b7ad5dbabb474d60fc1ec097e74c4008d", "submitted_at": "2024-04-11T17:56:05Z", "updated_at": "2024-05-30T08:55:12Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2404.07972v2" }, "canonical_source_url": "https://arxiv.org/abs/2404.07972", "catalog_id": "arxiv:2404.07972", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-04-11", "identifiers": { "arxiv": "2404.07972" }, "publication": { "authors": [ "Tianbao Xie", "Danyang Zhang", "Jixuan Chen", "Xiaochuan Li", "Siheng Zhao", "Ruisheng Cao", "Toh J. Hua", "Zhoujun Cheng", "Dongchan Shin", "Fangyu Lei", "Yitao Liu", "Yiheng Xu", "Shuyan Zhou", "Silvio Savarese", "Caiming Xiong", "Victor Zhong", "Tao Yu" ], "date_basis": "Explicit publisher citation_publication_date: 2024-12-16. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 12, 16 ], "doi": "10.52202/079017-1650", "event_dates": null, "evidence_urls": [ "https://proceedings.neurips.cc/paper_files/paper/2024/hash/5d413e48f84dc61244b6be550f1cd8f5-Abstract-Datasets_and_Benchmarks_Track.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "52040-52094", "pdf_url": "https://proceedings.neurips.cc/paper_files/paper/2024/file/5d413e48f84dc61244b6be550f1cd8f5-Paper-Datasets_and_Benchmarks_Track.pdf", "publisher": null, "series": null, "title": "OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments", "type": "conference-paper", "url": "https://proceedings.neurips.cc/paper_files/paper/2024/hash/5d413e48f84dc61244b6be550f1cd8f5-Abstract-Datasets_and_Benchmarks_Track.html", "venue": "Advances in Neural Information Processing Systems", "volume": "37" }, "record_updated_at": "2024-05-30" }, "genre": "preprint", "id": "arxiv:2404.07972", "issued": { "date-parts": [ [ 2024, 4, 11 ] ] }, "note": "First published: 2024-04-11; Recorded arXiv version: v2; Recorded version updated: 2024-05-30T08:55:12Z.", "title": "OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2404.08860v3", "archive": "arXiv", "archive_location": "2404.08860", "author": [ { "literal": "Lei Ding" }, { "literal": "Jeshwanth Bheemanpally" }, { "literal": "Yi Zhang" } ], "citation-key": "arxiv_2404_08860", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b3a030a53a942eacfabafc64af67591fd0a85af66291c5d1b48baea14556e8aa", "submitted_at": "2024-04-13T00:20:09Z", "updated_at": "2024-07-09T01:14:21Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2404.08860v3" }, "canonical_source_url": "https://arxiv.org/abs/2404.08860", "catalog_id": "arxiv:2404.08860", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-13", "identifiers": { "arxiv": "2404.08860" }, "record_updated_at": "2024-07-09" }, "genre": "preprint", "id": "arxiv:2404.08860", "issued": { "date-parts": [ [ 2024, 4, 13 ] ] }, "note": "First published: 2024-04-13; Recorded arXiv version: v3; Recorded version updated: 2024-07-09T01:14:21Z.", "title": "Enhancing Mobile \"How-to\" Queries with Automated Search Results Verification and Reranking", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2404.08948v4", "archive": "arXiv", "archive_location": "2404.08948", "author": [ { "literal": "Chenhui Cui" }, { "literal": "Tao Li" }, { "literal": "Junjie Wang" }, { "literal": "Chunyang Chen" }, { "literal": "Dave Towey" }, { "literal": "Rubing Huang" } ], "citation-key": "arxiv_2404_08948", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a844cdd06a32b8e89f362b7e924b36e27e72c1f1def81ed0b934f761ffcaba78", "submitted_at": "2024-04-13T09:56:50Z", "updated_at": "2026-06-25T14:05:55Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2404.08948v4" }, "canonical_source_url": "https://arxiv.org/abs/2404.08948", "catalog_id": "arxiv:2404.08948", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-13", "identifiers": { "arxiv": "2404.08948" }, "record_updated_at": "2026-06-25" }, "genre": "preprint", "id": "arxiv:2404.08948", "issued": { "date-parts": [ [ 2024, 4, 13 ] ] }, "note": "First published: 2024-04-13; Recorded arXiv version: v4; Recorded version updated: 2026-06-25T14:05:55Z.", "title": "Large Language Models for Mobile GUI Text Input Generation: An Empirical Study", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2404.09992v2", "archive": "arXiv", "archive_location": "2404.09992", "author": [ { "literal": "Shulin Tian" }, { "literal": "Ziniu Zhang" }, { "literal": "Liangyu Chen" }, { "literal": "Ziwei Liu" } ], "citation-key": "arxiv_2404_09992", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "667b1f5053f3ae6ea712999932d5e9ee2f42e402cc72e713048d0790f5931976", "submitted_at": "2024-04-15T17:59:50Z", "updated_at": "2025-06-29T05:24:54Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2404.09992v2" }, "canonical_source_url": "https://arxiv.org/abs/2404.09992", "catalog_id": "arxiv:2404.09992", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-04-15", "identifiers": { "arxiv": "2404.09992" }, "publication": { "authors": [ "Shulin Tian", "Ziniu Zhang", "Liang-Yu Chen", "Ziwei Liu" ], "date_basis": "Explicit publisher citation_publication_date: 2025/7. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 7 ], "doi": "10.18653/v1/2025.findings-acl.703", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.findings-acl.703/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "13682-13697", "pdf_url": "https://aclanthology.org/2025.findings-acl.703.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "MMInA: Benchmarking Multihop Multimodal Internet Agents", "type": "conference-paper", "url": "https://aclanthology.org/2025.findings-acl.703/", "venue": "Findings of the Association for Computational Linguistics: ACL 2025", "volume": null }, "record_updated_at": "2025-06-29" }, "genre": "preprint", "id": "arxiv:2404.09992", "issued": { "date-parts": [ [ 2024, 4, 15 ] ] }, "note": "First published: 2024-04-15; Recorded arXiv version: v2; Recorded version updated: 2025-06-29T05:24:54Z.", "title": "MMInA: Benchmarking Multihop Multimodal Internet Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2404.16054v2", "archive": "arXiv", "archive_location": "2404.16054", "author": [ { "literal": "Li Zhang" }, { "literal": "Shihe Wang" }, { "literal": "Xianqing Jia" }, { "literal": "Zhihan Zheng" }, { "literal": "Yunhe Yan" }, { "literal": "Longxi Gao" }, { "literal": "Yuanchun Li" }, { "literal": "Mengwei Xu" } ], "citation-key": "arxiv_2404_16054", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "28c3ad086106ec4199dcc9da2b8aa3b20eac9347d9354b9dfb373e50ea4c3f50", "submitted_at": "2024-04-12T15:39:09Z", "updated_at": "2024-08-02T13:49:32Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2404.16054v2" }, "canonical_source_url": "https://arxiv.org/abs/2404.16054", "catalog_id": "arxiv:2404.16054", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-12", "identifiers": { "arxiv": "2404.16054" }, "record_updated_at": "2024-08-02" }, "genre": "preprint", "id": "arxiv:2404.16054", "issued": { "date-parts": [ [ 2024, 4, 12 ] ] }, "note": "First published: 2024-04-12; Recorded arXiv version: v2; Recorded version updated: 2024-08-02T13:49:32Z.", "title": "LlamaTouch: A Faithful and Scalable Testbed for Mobile UI Task Automation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2404.16660v3", "archive": "arXiv", "archive_location": "2404.16660", "author": [ { "literal": "Juyong Lee" }, { "literal": "Taywon Min" }, { "literal": "Minyong An" }, { "literal": "Dongyoon Hahm" }, { "literal": "Haeone Lee" }, { "literal": "Changyeon Kim" }, { "literal": "Kimin Lee" } ], "citation-key": "arxiv_2404_16660", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "45c13064bbfbcf153566b86c8969f4de5b5e87dc133342c595b3b58bde3bb615", "submitted_at": "2024-04-25T14:56:32Z", "updated_at": "2025-07-21T02:55:09Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2404.16660v3" }, "canonical_source_url": "https://arxiv.org/abs/2404.16660", "catalog_id": "arxiv:2404.16660", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-25", "identifiers": { "arxiv": "2404.16660" }, "record_updated_at": "2025-07-21" }, "genre": "preprint", "id": "arxiv:2404.16660", "issued": { "date-parts": [ [ 2024, 4, 25 ] ] }, "note": "First published: 2024-04-25; Recorded arXiv version: v3; Recorded version updated: 2025-07-21T02:55:09Z.", "title": "Benchmarking Mobile Device Control Agents across Diverse Configurations", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2404.18074v3", "archive": "arXiv", "archive_location": "2404.18074", "author": [ { "literal": "Zirui Song" }, { "literal": "Yaohang Li" }, { "literal": "Meng Fang" }, { "literal": "Yanda Li" }, { "literal": "Zhenhao Chen" }, { "literal": "Zecheng Shi" }, { "literal": "Yuan Huang" }, { "literal": "Xiuying Chen" }, { "literal": "Ling Chen" } ], "citation-key": "arxiv_2404_18074", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5e531ec2ef9e5d90586b8f3052e16d41da95114b7382047d0a02fd671a0db355", "submitted_at": "2024-04-28T05:33:15Z", "updated_at": "2025-03-23T13:04:57Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2404.18074v3" }, "canonical_source_url": "https://arxiv.org/abs/2404.18074", "catalog_id": "arxiv:2404.18074", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-04-28", "identifiers": { "arxiv": "2404.18074" }, "record_updated_at": "2025-03-23" }, "genre": "preprint", "id": "arxiv:2404.18074", "issued": { "date-parts": [ [ 2024, 4, 28 ] ] }, "note": "First published: 2024-04-28; Recorded arXiv version: v3; Recorded version updated: 2025-03-23T13:04:57Z.", "title": "MMAC-Copilot: Multi-modal Agent Collaboration Operating Copilot", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2405.00516v1", "archive": "arXiv", "archive_location": "2405.00516", "author": [ { "literal": "Lucas-Andreï Thil" }, { "literal": "Mirela Popa" }, { "literal": "Gerasimos Spanakis" } ], "citation-key": "arxiv_2405_00516", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3498a7bc289c6537ab69780a56a9fd0a2bb697fb74c077949824c4fc018762c6", "submitted_at": "2024-05-01T13:51:45Z", "updated_at": "2024-05-01T13:51:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2405.00516v1" }, "canonical_source_url": "https://arxiv.org/abs/2405.00516", "catalog_id": "arxiv:2405.00516", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-05-01", "identifiers": { "arxiv": "2405.00516" }, "record_updated_at": "2024-05-01" }, "genre": "preprint", "id": "arxiv:2405.00516", "issued": { "date-parts": [ [ 2024, 5, 1 ] ] }, "note": "First published: 2024-05-01; Recorded arXiv version: v1; Recorded version updated: 2024-05-01T13:51:45Z.", "title": "Navigating WebAI: Training Agents to Complete Web Tasks with Large Language Models and Reinforcement Learning", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2405.04497v2", "archive": "arXiv", "archive_location": "2405.04497", "author": [ { "literal": "Kihoon Son" }, { "literal": "Jinhyeon Kwon" }, { "literal": "DaEun Choi" }, { "literal": "Tae Soo Kim" }, { "literal": "Young-Ho Kim" }, { "literal": "Sangdoo Yun" }, { "literal": "Juho Kim" } ], "citation-key": "arxiv_2405_04497", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a358d417d8bef89a01e799d88564e3ede4678ebda4950e87c96cf1e1954b85e1", "submitted_at": "2024-05-07T17:10:31Z", "updated_at": "2024-05-08T05:44:21Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2405.04497v2" }, "canonical_source_url": "https://arxiv.org/abs/2405.04497", "catalog_id": "arxiv:2405.04497", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-05-07", "identifiers": { "arxiv": "2405.04497" }, "record_updated_at": "2024-05-08" }, "genre": "preprint", "id": "arxiv:2405.04497", "issued": { "date-parts": [ [ 2024, 5, 7 ] ] }, "note": "First published: 2024-05-07; Recorded arXiv version: v2; Recorded version updated: 2024-05-08T05:44:21Z.", "title": "Unveiling Disparities in Web Task Handling Between Human and Web Agent", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2405.11120v1", "archive": "arXiv", "archive_location": "2405.11120", "author": [ { "literal": "William E Bishop" }, { "literal": "Alice Li" }, { "literal": "Christopher Rawles" }, { "literal": "Oriana Riva" } ], "citation-key": "arxiv_2405_11120", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "81f62326dddc5c512fb05472eb1f387064c0a5aad77453e9d2e85415bb627fa9", "submitted_at": "2024-05-17T23:27:33Z", "updated_at": "2024-05-17T23:27:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2405.11120v1" }, "canonical_source_url": "https://arxiv.org/abs/2405.11120", "catalog_id": "arxiv:2405.11120", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-05-17", "identifiers": { "arxiv": "2405.11120" }, "record_updated_at": "2024-05-17" }, "genre": "preprint", "id": "arxiv:2405.11120", "issued": { "date-parts": [ [ 2024, 5, 17 ] ] }, "note": "First published: 2024-05-17; Recorded arXiv version: v1; Recorded version updated: 2024-05-17T23:27:33Z.", "title": "Latent State Estimation Helps UI Agents to Reason", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2405.14573v5", "archive": "arXiv", "archive_location": "2405.14573", "author": [ { "literal": "Christopher Rawles" }, { "literal": "Sarah Clinckemaillie" }, { "literal": "Yifan Chang" }, { "literal": "Jonathan Waltz" }, { "literal": "Gabrielle Lau" }, { "literal": "Marybeth Fair" }, { "literal": "Alice Li" }, { "literal": "William Bishop" }, { "literal": "Wei Li" }, { "literal": "Folawiyo Campbell-Ajala" }, { "literal": "Daniel Toyama" }, { "literal": "Robert Berry" }, { "literal": "Divya Tyamagundlu" }, { "literal": "Timothy Lillicrap" }, { "literal": "Oriana Riva" } ], "citation-key": "arxiv_2405_14573", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a33b8b0a03c0dc7ac3f699c34193a526d20791bfedfdde2a2fb89604959e08c2", "submitted_at": "2024-05-23T13:48:54Z", "updated_at": "2025-04-06T20:37:50Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2405.14573v5" }, "canonical_source_url": "https://arxiv.org/abs/2405.14573", "catalog_id": "arxiv:2405.14573", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-05-23", "identifiers": { "arxiv": "2405.14573" }, "publication": { "authors": [ "Chris Rawles", "Sarah Clinckemaillie", "Yifan Chang", "Jonathan Waltz", "Gabrielle Lau", "Marybeth Fair", "Alice Li", "William Bishop", "Wei Li", "Folawiyo Campbell-Ajala", "Daniel Toyama", "Robert Berry", "Divya Tyamagundlu", "Timothy Lillicrap", "Oriana Riva" ], "date_basis": "Explicit publisher citation_publication_date: 2025-05-01. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 5, 1 ], "doi": null, "event_dates": null, "evidence_urls": [ "https://proceedings.iclr.cc/paper_files/paper/2025/hash/01a83bc2f2732a58e6aa731e659e7101-Abstract-Conference.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "406-441", "pdf_url": "https://proceedings.iclr.cc/paper_files/paper/2025/file/01a83bc2f2732a58e6aa731e659e7101-Paper-Conference.pdf", "publisher": null, "series": null, "title": "AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents", "type": "conference-paper", "url": "https://proceedings.iclr.cc/paper_files/paper/2025/hash/01a83bc2f2732a58e6aa731e659e7101-Abstract-Conference.html", "venue": "International Conference on Learning Representations", "volume": "2025" }, "record_updated_at": "2025-04-06" }, "genre": "preprint", "id": "arxiv:2405.14573", "issued": { "date-parts": [ [ 2024, 5, 23 ] ] }, "note": "First published: 2024-05-23; Recorded arXiv version: v5; Recorded version updated: 2025-04-06T20:37:50Z.", "title": "AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2405.20309v2", "archive": "arXiv", "archive_location": "2405.20309", "author": [ { "literal": "Ajay Patel" }, { "literal": "Markus Hofmarcher" }, { "literal": "Claudiu Leoveanu-Condrei" }, { "literal": "Marius-Constantin Dinu" }, { "literal": "Chris Callison-Burch" }, { "literal": "Sepp Hochreiter" } ], "citation-key": "arxiv_2405_20309", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "77b292052a2ab4e8db61606ffe1da8bd5e63b3ce8c84b2e0e4b7a2fdc2f926ca", "submitted_at": "2024-05-30T17:52:36Z", "updated_at": "2024-10-01T21:28:29Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2405.20309v2" }, "canonical_source_url": "https://arxiv.org/abs/2405.20309", "catalog_id": "arxiv:2405.20309", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-05-30", "identifiers": { "arxiv": "2405.20309" }, "record_updated_at": "2024-10-01" }, "genre": "preprint", "id": "arxiv:2405.20309", "issued": { "date-parts": [ [ 2024, 5, 30 ] ] }, "note": "First published: 2024-05-30; Recorded arXiv version: v2; Recorded version updated: 2024-10-01T21:28:29Z.", "title": "Large Language Models Can Self-Improve At Web Agent Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2406.01623v1", "archive": "arXiv", "archive_location": "2406.01623", "author": [ { "literal": "Eric Li" }, { "literal": "Jim Waldo" } ], "citation-key": "arxiv_2406_01623", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b9a728edf0b834adb72abfc50e13d5b2b52f532f444b3e8af34effd3d86beb1b", "submitted_at": "2024-06-01T00:32:26Z", "updated_at": "2024-06-01T00:32:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2406.01623v1" }, "canonical_source_url": "https://arxiv.org/abs/2406.01623", "catalog_id": "arxiv:2406.01623", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-01", "identifiers": { "arxiv": "2406.01623" }, "record_updated_at": "2024-06-01" }, "genre": "preprint", "id": "arxiv:2406.01623", "issued": { "date-parts": [ [ 2024, 6, 1 ] ] }, "note": "First published: 2024-06-01; Recorded arXiv version: v1; Recorded version updated: 2024-06-01T00:32:26Z.", "title": "WebSuite: Systematically Evaluating Why Web Agents Fail", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2406.03679v6", "archive": "arXiv", "archive_location": "2406.03679", "author": [ { "literal": "Wei Li" }, { "literal": "William Bishop" }, { "literal": "Alice Li" }, { "literal": "Chris Rawles" }, { "literal": "Folawiyo Campbell-Ajala" }, { "literal": "Divya Tyamagundlu" }, { "literal": "Oriana Riva" } ], "citation-key": "arxiv_2406_03679", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6eb3209d8e2a431987fe19f01cd0818f9fb490e684d7369e7d3002dc388ff271", "submitted_at": "2024-06-06T01:49:29Z", "updated_at": "2024-11-13T16:42:22Z", "version": 6, "versioned_url": "https://arxiv.org/abs/2406.03679v6" }, "canonical_source_url": "https://arxiv.org/abs/2406.03679", "catalog_id": "arxiv:2406.03679", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-06-06", "identifiers": { "arxiv": "2406.03679" }, "publication": { "authors": [ "Wei Li", "William Bishop", "Alice Li", "Chris Rawles", "Folawiyo Campbell-Ajala", "Divya Tyamagundlu", "Oriana Riva" ], "date_basis": "Explicit publisher citation_publication_date: 2024-12-16. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 12, 16 ], "doi": "10.52202/079017-2925", "event_dates": null, "evidence_urls": [ "https://proceedings.neurips.cc/paper_files/paper/2024/hash/a79f3ef3b445fd4659f44648f7ea8ffd-Abstract-Datasets_and_Benchmarks_Track.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "92130-92154", "pdf_url": "https://proceedings.neurips.cc/paper_files/paper/2024/file/a79f3ef3b445fd4659f44648f7ea8ffd-Paper-Datasets_and_Benchmarks_Track.pdf", "publisher": null, "series": null, "title": "On the Effects of Data Scale on UI Control Agents", "type": "conference-paper", "url": "https://proceedings.neurips.cc/paper_files/paper/2024/hash/a79f3ef3b445fd4659f44648f7ea8ffd-Abstract-Datasets_and_Benchmarks_Track.html", "venue": "Advances in Neural Information Processing Systems", "volume": "37" }, "record_updated_at": "2024-11-13" }, "genre": "preprint", "id": "arxiv:2406.03679", "issued": { "date-parts": [ [ 2024, 6, 6 ] ] }, "note": "First published: 2024-06-06; Recorded arXiv version: v6; Recorded version updated: 2024-11-13T16:42:22Z.", "title": "On the Effects of Data Scale on UI Control Agents", "type": "article", "version": "6" }, { "URL": "https://arxiv.org/abs/2406.08184v1", "archive": "arXiv", "archive_location": "2406.08184", "author": [ { "literal": "Luyuan Wang" }, { "literal": "Yongyu Deng" }, { "literal": "Yiwei Zha" }, { "literal": "Guodong Mao" }, { "literal": "Qinmin Wang" }, { "literal": "Tianchen Min" }, { "literal": "Wei Chen" }, { "literal": "Shoufa Chen" } ], "citation-key": "arxiv_2406_08184", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "637843b1b8634e608bed71fc0f3df205ae218d1b0b5550ce480046c67bb14b0f", "submitted_at": "2024-06-12T13:14:50Z", "updated_at": "2024-06-12T13:14:50Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2406.08184v1" }, "canonical_source_url": "https://arxiv.org/abs/2406.08184", "catalog_id": "arxiv:2406.08184", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-12", "identifiers": { "arxiv": "2406.08184" }, "record_updated_at": "2024-06-12" }, "genre": "preprint", "id": "arxiv:2406.08184", "issued": { "date-parts": [ [ 2024, 6, 12 ] ] }, "note": "First published: 2024-06-12; Recorded arXiv version: v1; Recorded version updated: 2024-06-12T13:14:50Z.", "title": "MobileAgentBench: An Efficient and User-Friendly Benchmark for Mobile LLM Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2406.08340v1", "archive": "arXiv", "archive_location": "2406.08340", "author": [ { "literal": "Shengcheng Yu" }, { "literal": "Chunrong Fang" }, { "literal": "Mingzhe Du" }, { "literal": "Zimin Ding" }, { "literal": "Zhenyu Chen" }, { "literal": "Zhendong Su" } ], "citation-key": "arxiv_2406_08340", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0061e9004afe11bf188164ea37de1acd15e1b95ae2fcad22044264a89e4636b3", "submitted_at": "2024-06-12T15:48:39Z", "updated_at": "2024-06-12T15:48:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2406.08340v1" }, "canonical_source_url": "https://arxiv.org/abs/2406.08340", "catalog_id": "arxiv:2406.08340", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-12", "identifiers": { "arxiv": "2406.08340" }, "record_updated_at": "2024-06-12" }, "genre": "preprint", "id": "arxiv:2406.08340", "issued": { "date-parts": [ [ 2024, 6, 12 ] ] }, "note": "First published: 2024-06-12; Recorded arXiv version: v1; Recorded version updated: 2024-06-12T15:48:39Z.", "title": "Practical, Automated Scenario-based Mobile App Testing", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2406.09187v3", "archive": "arXiv", "archive_location": "2406.09187", "author": [ { "literal": "Zhen Xiang" }, { "literal": "Linzhi Zheng" }, { "literal": "Yanjie Li" }, { "literal": "Junyuan Hong" }, { "literal": "Qinbin Li" }, { "literal": "Han Xie" }, { "literal": "Jiawei Zhang" }, { "literal": "Zidi Xiong" }, { "literal": "Chulin Xie" }, { "literal": "Carl Yang" }, { "literal": "Dawn Song" }, { "literal": "Bo Li" } ], "citation-key": "arxiv_2406_09187", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "50a8ea492d3f1c2ff2bd00ebfb18e19077bacc96b44c751d985bfe0f7bef0247", "submitted_at": "2024-06-13T14:49:26Z", "updated_at": "2025-05-29T03:09:05Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2406.09187v3" }, "canonical_source_url": "https://arxiv.org/abs/2406.09187", "catalog_id": "arxiv:2406.09187", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-06-13", "identifiers": { "arxiv": "2406.09187" }, "record_updated_at": "2025-05-29" }, "genre": "preprint", "id": "arxiv:2406.09187", "issued": { "date-parts": [ [ 2024, 6, 13 ] ] }, "note": "First published: 2024-06-13; Recorded arXiv version: v3; Recorded version updated: 2025-05-29T03:09:05Z.", "title": "GuardAgent: Safeguard LLM Agents by a Guard Agent via Knowledge-Enabled Reasoning", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2406.10227v1", "archive": "arXiv", "archive_location": "2406.10227", "author": [ { "literal": "Kevin Qinghong Lin" }, { "literal": "Linjie Li" }, { "literal": "Difei Gao" }, { "literal": "Qinchen WU" }, { "literal": "Mingyi Yan" }, { "literal": "Zhengyuan Yang" }, { "literal": "Lijuan Wang" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2406_10227", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "66b5fb1e839490df80e9750fdc049f7fe3d2d4686c4090d78ea01bc542fe55b0", "submitted_at": "2024-06-14T17:59:08Z", "updated_at": "2024-06-14T17:59:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2406.10227v1" }, "canonical_source_url": "https://arxiv.org/abs/2406.10227", "catalog_id": "arxiv:2406.10227", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-06-14", "identifiers": { "arxiv": "2406.10227" }, "publication": { "authors": [ "Kevin Q. Lin", "Linjie Li", "Difei Gao", "Qinchen Wu", "Mingyi Yan", "Zhengyuan Yang", "Lijuan Wang", "Mike Z. Shou" ], "date_basis": "Explicit publisher citation_publication_date: 2024-12-16. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 12, 16 ], "doi": "10.52202/079017-2214", "event_dates": null, "evidence_urls": [ "https://proceedings.neurips.cc/paper_files/paper/2024/hash/804e757b7d7043c26701c3a313032101-Abstract-Datasets_and_Benchmarks_Track.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "69329-69360", "pdf_url": "https://proceedings.neurips.cc/paper_files/paper/2024/file/804e757b7d7043c26701c3a313032101-Paper-Datasets_and_Benchmarks_Track.pdf", "publisher": null, "series": null, "title": "VideoGUI: A Benchmark for GUI Automation from Instructional Videos", "type": "conference-paper", "url": "https://proceedings.neurips.cc/paper_files/paper/2024/hash/804e757b7d7043c26701c3a313032101-Abstract-Datasets_and_Benchmarks_Track.html", "venue": "Advances in Neural Information Processing Systems", "volume": "37" }, "record_updated_at": "2024-06-14" }, "genre": "preprint", "id": "arxiv:2406.10227", "issued": { "date-parts": [ [ 2024, 6, 14 ] ] }, "note": "First published: 2024-06-14; Recorded arXiv version: v1; Recorded version updated: 2024-06-14T17:59:08Z.", "title": "VideoGUI: A Benchmark for GUI Automation from Instructional Videos", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2406.10819v2", "archive": "arXiv", "archive_location": "2406.10819", "author": [ { "literal": "Dongping Chen" }, { "literal": "Yue Huang" }, { "literal": "Siyuan Wu" }, { "literal": "Jingyu Tang" }, { "literal": "Liuyi Chen" }, { "literal": "Yilin Bai" }, { "literal": "Zhigang He" }, { "literal": "Chenlong Wang" }, { "literal": "Huichi Zhou" }, { "literal": "Yiqiang Li" }, { "literal": "Tianshuo Zhou" }, { "literal": "Yue Yu" }, { "literal": "Chujie Gao" }, { "literal": "Qihui Zhang" }, { "literal": "Yi Gui" }, { "literal": "Zhen Li" }, { "literal": "Yao Wan" }, { "literal": "Pan Zhou" }, { "literal": "Jianfeng Gao" }, { "literal": "Lichao Sun" } ], "citation-key": "arxiv_2406_10819", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9997c3ab43f0e3e69ad4dcbe15257ea4392d36fa3c82f90c6b7730667a067f99", "submitted_at": "2024-06-16T06:56:53Z", "updated_at": "2025-03-24T11:46:14Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2406.10819v2" }, "canonical_source_url": "https://arxiv.org/abs/2406.10819", "catalog_id": "arxiv:2406.10819", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-16", "identifiers": { "arxiv": "2406.10819" }, "record_updated_at": "2025-03-24" }, "genre": "preprint", "id": "arxiv:2406.10819", "issued": { "date-parts": [ [ 2024, 6, 16 ] ] }, "note": "First published: 2024-06-16; Recorded arXiv version: v2; Recorded version updated: 2025-03-24T11:46:14Z.", "title": "GUI-World: A Video Benchmark and Dataset for Multimodal GUI-oriented Understanding", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2406.12373v3", "archive": "arXiv", "archive_location": "2406.12373", "author": [ { "literal": "Yichen Pan" }, { "literal": "Dehan Kong" }, { "literal": "Sida Zhou" }, { "literal": "Cheng Cui" }, { "literal": "Yifei Leng" }, { "literal": "Bing Jiang" }, { "literal": "Hangyu Liu" }, { "literal": "Yanyi Shang" }, { "literal": "Shuyan Zhou" }, { "literal": "Tongshuang Wu" }, { "literal": "Zhengyang Wu" } ], "citation-key": "arxiv_2406_12373", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c5d0fbea7a4813a1da85c9ecb237c2115cc183b4c9e6fd0cbfd63764b0d418e3", "submitted_at": "2024-06-18T07:58:33Z", "updated_at": "2024-07-16T06:19:21Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2406.12373v3" }, "canonical_source_url": "https://arxiv.org/abs/2406.12373", "catalog_id": "arxiv:2406.12373", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-18", "identifiers": { "arxiv": "2406.12373" }, "record_updated_at": "2024-07-16" }, "genre": "preprint", "id": "arxiv:2406.12373", "issued": { "date-parts": [ [ 2024, 6, 18 ] ] }, "note": "First published: 2024-06-18; Recorded arXiv version: v3; Recorded version updated: 2024-07-16T06:19:21Z.", "title": "WebCanvas: Benchmarking Web Agents in Online Environments", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2406.12814v3", "archive": "arXiv", "archive_location": "2406.12814", "author": [ { "literal": "Chen Henry Wu" }, { "literal": "Rishi Shah" }, { "literal": "Jing Yu Koh" }, { "literal": "Ruslan Salakhutdinov" }, { "literal": "Daniel Fried" }, { "literal": "Aditi Raghunathan" } ], "citation-key": "arxiv_2406_12814", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7cc1c1ac93f8e328d33accf7cdaa1d65487c4f7abec15234b9e596a501947550", "submitted_at": "2024-06-18T17:32:48Z", "updated_at": "2025-02-04T20:02:17Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2406.12814v3" }, "canonical_source_url": "https://arxiv.org/abs/2406.12814", "catalog_id": "arxiv:2406.12814", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-06-18", "identifiers": { "arxiv": "2406.12814" }, "publication": { "authors": [ "Chen Wu", "Rishi Shah", "Jing Yu Koh", "Russ Salakhutdinov", "Daniel Fried", "Aditi Raghunathan" ], "date_basis": "Explicit publisher citation_publication_date: 2025-05-01. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 5, 1 ], "doi": null, "event_dates": null, "evidence_urls": [ "https://proceedings.iclr.cc/paper_files/paper/2025/hash/460a1d8eac34125dad453b28d6d64446-Abstract-Conference.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity. The normalized publisher abstract matches the retained arXiv abstract SHA-256.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "28362-28383", "pdf_url": "https://proceedings.iclr.cc/paper_files/paper/2025/file/460a1d8eac34125dad453b28d6d64446-Paper-Conference.pdf", "publisher": null, "series": null, "title": "Dissecting Adversarial Robustness of Multimodal LM Agents", "type": "conference-paper", "url": "https://proceedings.iclr.cc/paper_files/paper/2025/hash/460a1d8eac34125dad453b28d6d64446-Abstract-Conference.html", "venue": "International Conference on Learning Representations", "volume": "2025" }, "record_updated_at": "2025-02-04" }, "genre": "preprint", "id": "arxiv:2406.12814", "issued": { "date-parts": [ [ 2024, 6, 18 ] ] }, "note": "First published: 2024-06-18; Recorded arXiv version: v3; Recorded version updated: 2025-02-04T20:02:17Z.", "title": "Dissecting Adversarial Robustness of Multimodal LM Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2406.13264v2", "archive": "arXiv", "archive_location": "2406.13264", "author": [ { "literal": "Michael Wornow" }, { "literal": "Avanika Narayan" }, { "literal": "Ben Viggiano" }, { "literal": "Ishan S. Khare" }, { "literal": "Tathagat Verma" }, { "literal": "Tibor Thompson" }, { "literal": "Miguel Angel Fuentes Hernandez" }, { "literal": "Sudharsan Sundar" }, { "literal": "Chloe Trujillo" }, { "literal": "Krrish Chawla" }, { "literal": "Rongfei Lu" }, { "literal": "Justin Shen" }, { "literal": "Divya Nagaraj" }, { "literal": "Joshua Martinez" }, { "literal": "Vardhan Agrawal" }, { "literal": "Althea Hudson" }, { "literal": "Nigam H. Shah" }, { "literal": "Christopher Re" } ], "citation-key": "arxiv_2406_13264", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a7e7d00030ea8c77b9ed6f950d62f2bd417d3870f24b2f4589d8068c9484e328", "submitted_at": "2024-06-19T06:50:15Z", "updated_at": "2024-10-11T00:06:31Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2406.13264v2" }, "canonical_source_url": "https://arxiv.org/abs/2406.13264", "catalog_id": "arxiv:2406.13264", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-19", "identifiers": { "arxiv": "2406.13264" }, "record_updated_at": "2024-10-11" }, "genre": "preprint", "id": "arxiv:2406.13264", "issued": { "date-parts": [ [ 2024, 6, 19 ] ] }, "note": "First published: 2024-06-19; Recorded arXiv version: v2; Recorded version updated: 2024-10-11T00:06:31Z.", "title": "WONDERBREAD: A Benchmark for Evaluating Multimodal Foundation Models on Business Process Management Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2406.14250v3", "archive": "arXiv", "archive_location": "2406.14250", "author": [ { "literal": "Ke Wang" }, { "literal": "Tianyu Xia" }, { "literal": "Zhangxuan Gu" }, { "literal": "Yi Zhao" }, { "literal": "Shuheng Shen" }, { "literal": "Changhua Meng" }, { "literal": "Weiqiang Wang" }, { "literal": "Ke Xu" } ], "citation-key": "arxiv_2406_14250", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e4d5a46d2141ca59fcf221bbe618601cb7cce29c1b067ae60712e53e907095bd", "submitted_at": "2024-06-20T12:22:05Z", "updated_at": "2024-07-01T08:37:14Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2406.14250v3" }, "canonical_source_url": "https://arxiv.org/abs/2406.14250", "catalog_id": "arxiv:2406.14250", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-20", "identifiers": { "arxiv": "2406.14250" }, "record_updated_at": "2024-07-01" }, "genre": "preprint", "id": "arxiv:2406.14250", "issued": { "date-parts": [ [ 2024, 6, 20 ] ] }, "note": "First published: 2024-06-20; Recorded arXiv version: v3; Recorded version updated: 2024-07-01T08:37:14Z.", "title": "E-ANT: A Large-Scale Dataset for Efficient Automatic GUI NavigaTion", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2406.14314v3", "archive": "arXiv", "archive_location": "2406.14314", "author": [ { "literal": "Omri Berkovitch" }, { "literal": "Sapir Caduri" }, { "literal": "Noam Kahlon" }, { "literal": "Anatoly Efros" }, { "literal": "Avi Caciularu" }, { "literal": "Ido Dagan" } ], "citation-key": "arxiv_2406_14314", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a0bc798b69e59d1ec0ca8ecb0fab7a7ee6594f4d656ca4212b19be7e3085c0c5", "submitted_at": "2024-06-20T13:46:10Z", "updated_at": "2025-03-03T15:47:10Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2406.14314v3" }, "canonical_source_url": "https://arxiv.org/abs/2406.14314", "catalog_id": "arxiv:2406.14314", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-06-20", "identifiers": { "arxiv": "2406.14314" }, "record_updated_at": "2025-03-03" }, "genre": "preprint", "id": "arxiv:2406.14314", "issued": { "date-parts": [ [ 2024, 6, 20 ] ] }, "note": "First published: 2024-06-20; Recorded arXiv version: v3; Recorded version updated: 2025-03-03T15:47:10Z.", "title": "Identifying User Goals from UI Trajectories", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2407.00993v1", "archive": "arXiv", "archive_location": "2407.00993", "author": [ { "literal": "Shihan Deng" }, { "literal": "Weikai Xu" }, { "literal": "Hongda Sun" }, { "literal": "Wei Liu" }, { "literal": "Tao Tan" }, { "literal": "Jianfeng Liu" }, { "literal": "Ang Li" }, { "literal": "Jian Luan" }, { "literal": "Bin Wang" }, { "literal": "Rui Yan" }, { "literal": "Shuo Shang" } ], "citation-key": "arxiv_2407_00993", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "90ce9900644c21004e0ba9e749e10d94238d2447d0e84bd7205ff55dcbd243f4", "submitted_at": "2024-07-01T06:10:01Z", "updated_at": "2024-07-01T06:10:01Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2407.00993v1" }, "canonical_source_url": "https://arxiv.org/abs/2407.00993", "catalog_id": "arxiv:2407.00993", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-07-01", "identifiers": { "arxiv": "2407.00993" }, "record_updated_at": "2024-07-01" }, "genre": "preprint", "id": "arxiv:2407.00993", "issued": { "date-parts": [ [ 2024, 7, 1 ] ] }, "note": "First published: 2024-07-01; Recorded arXiv version: v1; Recorded version updated: 2024-07-01T06:10:01Z.", "title": "Mobile-Bench: An Evaluation Benchmark for LLM-based Mobile Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2407.01476v4", "archive": "arXiv", "archive_location": "2407.01476", "author": [ { "literal": "Jing Yu Koh" }, { "literal": "Stephen McAleer" }, { "literal": "Daniel Fried" }, { "literal": "Ruslan Salakhutdinov" } ], "citation-key": "arxiv_2407_01476", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ec7aaf2a34e38f72da652bb2d8350c951229c154e2bc368a927a79287b36c439", "submitted_at": "2024-07-01T17:07:55Z", "updated_at": "2026-02-08T15:06:40Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2407.01476v4" }, "canonical_source_url": "https://arxiv.org/abs/2407.01476", "catalog_id": "arxiv:2407.01476", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-07-01", "identifiers": { "arxiv": "2407.01476" }, "record_updated_at": "2026-02-08" }, "genre": "preprint", "id": "arxiv:2407.01476", "issued": { "date-parts": [ [ 2024, 7, 1 ] ] }, "note": "First published: 2024-07-01; Recorded arXiv version: v4; Recorded version updated: 2026-02-08T15:06:40Z.", "title": "Tree Search for Language Model Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2407.01511v4", "archive": "arXiv", "archive_location": "2407.01511", "author": [ { "literal": "Tianqi Xu" }, { "literal": "Linyao Chen" }, { "literal": "Dai-Jie Wu" }, { "literal": "Yanjun Chen" }, { "literal": "Zecheng Zhang" }, { "literal": "Xiang Yao" }, { "literal": "Zhiqiang Xie" }, { "literal": "Yongchao Chen" }, { "literal": "Shilong Liu" }, { "literal": "Bochen Qian" }, { "literal": "Anjie Yang" }, { "literal": "Zhaoxuan Jin" }, { "literal": "Jianbo Deng" }, { "literal": "Philip Torr" }, { "literal": "Bernard Ghanem" }, { "literal": "Guohao Li" } ], "citation-key": "arxiv_2407_01511", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "12e5464ea0fc62bd566f19d301b887826b7ff317cfc27c86fa0cda8d3de271d9", "submitted_at": "2024-07-01T17:55:04Z", "updated_at": "2025-07-20T13:42:07Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2407.01511v4" }, "canonical_source_url": "https://arxiv.org/abs/2407.01511", "catalog_id": "arxiv:2407.01511", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-07-01", "identifiers": { "arxiv": "2407.01511" }, "publication": { "authors": [ "Tianqi Xu", "Linyao Chen", "Dai-Jie Wu", "Yanjun Chen", "Zecheng Zhang", "Xiang Yao", "Zhiqiang Xie (谢志强)", "Yongchao Chen", "Shilong Liu", "Bochen Qian", "Anjie Yang", "Zhaoxuan Jin", "Jianbo Deng", "Philip Torr", "Bernard Ghanem", "Guohao Li" ], "date_basis": "Explicit publisher citation_publication_date: 2025/7. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 7 ], "doi": "10.18653/v1/2025.findings-acl.1113", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.findings-acl.1113/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "21607-21647", "pdf_url": "https://aclanthology.org/2025.findings-acl.1113.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents", "type": "conference-paper", "url": "https://aclanthology.org/2025.findings-acl.1113/", "venue": "Findings of the Association for Computational Linguistics: ACL 2025", "volume": null }, "record_updated_at": "2025-07-20" }, "genre": "preprint", "id": "arxiv:2407.01511", "issued": { "date-parts": [ [ 2024, 7, 1 ] ] }, "note": "First published: 2024-07-01; Recorded arXiv version: v4; Recorded version updated: 2025-07-20T13:42:07Z.", "title": "CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2407.01558v3", "archive": "arXiv", "archive_location": "2407.01558", "author": [ { "literal": "El Hassane Ettifouri" }, { "literal": "Jessica López Espejel" }, { "literal": "Laura Minkova" }, { "literal": "Tassnim Dardouri" }, { "literal": "Walid Dahhane" } ], "citation-key": "arxiv_2407_01558", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "01d3b3a013f3b10249b8197a8e18aa695bdc20febaa34f07128791f36afdcbac", "submitted_at": "2024-05-05T19:10:19Z", "updated_at": "2025-07-18T11:35:01Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2407.01558v3" }, "canonical_source_url": "https://arxiv.org/abs/2407.01558", "catalog_id": "arxiv:2407.01558", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-05-05", "identifiers": { "arxiv": "2407.01558" }, "record_updated_at": "2025-07-18" }, "genre": "preprint", "id": "arxiv:2407.01558", "issued": { "date-parts": [ [ 2024, 5, 5 ] ] }, "note": "First published: 2024-05-05; Recorded arXiv version: v3; Recorded version updated: 2025-07-18T11:35:01Z.", "title": "Visual Grounding Methods for Efficient Interaction with Desktop Graphical User Interfaces", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2407.03037v2", "archive": "arXiv", "archive_location": "2407.03037", "author": [ { "literal": "Zhe Liu" }, { "literal": "Cheng Li" }, { "literal": "Chunyang Chen" }, { "literal": "Junjie Wang" }, { "literal": "Mengzhuo Chen" }, { "literal": "Boyu Wu" }, { "literal": "Yawen Wang" }, { "literal": "Jun Hu" }, { "literal": "Qing Wang" } ], "citation-key": "arxiv_2407_03037", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "73b706ac6089a7d19836268437f3fea58d10b9d170ac596ac0d5a12f0d3a9668", "submitted_at": "2024-07-03T11:58:09Z", "updated_at": "2024-12-04T08:00:42Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2407.03037v2" }, "canonical_source_url": "https://arxiv.org/abs/2407.03037", "catalog_id": "arxiv:2407.03037", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-07-03", "identifiers": { "arxiv": "2407.03037" }, "record_updated_at": "2024-12-04" }, "genre": "preprint", "id": "arxiv:2407.03037", "issued": { "date-parts": [ [ 2024, 7, 3 ] ] }, "note": "First published: 2024-07-03; Recorded arXiv version: v2; Recorded version updated: 2024-12-04T08:00:42Z.", "title": "Seeing is Believing: Vision-driven Non-crash Functional Bug Detection for Mobile Apps", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2407.05291v2", "archive": "arXiv", "archive_location": "2407.05291", "author": [ { "literal": "Léo Boisvert" }, { "literal": "Megh Thakkar" }, { "literal": "Maxime Gasse" }, { "literal": "Massimo Caccia" }, { "literal": "Thibault Le Sellier De Chezelles" }, { "literal": "Quentin Cappart" }, { "literal": "Nicolas Chapados" }, { "literal": "Alexandre Lacoste" }, { "literal": "Alexandre Drouin" } ], "citation-key": "arxiv_2407_05291", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2b56a6208e4fa3523cae3fbae673affcd5fe3d6a9c19df482d27fe71d43ffef5", "submitted_at": "2024-07-07T07:15:49Z", "updated_at": "2025-02-05T21:50:07Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2407.05291v2" }, "canonical_source_url": "https://arxiv.org/abs/2407.05291", "catalog_id": "arxiv:2407.05291", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-07-07", "identifiers": { "arxiv": "2407.05291" }, "publication": { "authors": [ "Léo Boisvert", "Megh Thakkar", "Maxime Gasse", "Massimo Caccia", "Thibault L. De Chezelles", "Quentin Cappart", "Nicolas Chapados", "Alexandre Lacoste", "Alexandre Drouin" ], "date_basis": "Explicit publisher citation_publication_date: 2024-12-16. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 12, 16 ], "doi": "10.52202/079017-0195", "event_dates": null, "evidence_urls": [ "https://proceedings.neurips.cc/paper_files/paper/2024/hash/0b82662b6c32e887bb252a74d8cb2d5e-Abstract-Datasets_and_Benchmarks_Track.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "5996-6051", "pdf_url": "https://proceedings.neurips.cc/paper_files/paper/2024/file/0b82662b6c32e887bb252a74d8cb2d5e-Paper-Datasets_and_Benchmarks_Track.pdf", "publisher": null, "series": null, "title": "WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks", "type": "conference-paper", "url": "https://proceedings.neurips.cc/paper_files/paper/2024/hash/0b82662b6c32e887bb252a74d8cb2d5e-Abstract-Datasets_and_Benchmarks_Track.html", "venue": "Advances in Neural Information Processing Systems", "volume": "37" }, "record_updated_at": "2025-02-05" }, "genre": "preprint", "id": "arxiv:2407.05291", "issued": { "date-parts": [ [ 2024, 7, 7 ] ] }, "note": "First published: 2024-07-07; Recorded arXiv version: v2; Recorded version updated: 2025-02-05T21:50:07Z.", "title": "WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2407.08850v3", "archive": "arXiv", "archive_location": "2407.08850", "author": [ { "literal": "Peitong Duan" }, { "literal": "Chin-yi Chen" }, { "literal": "Gang Li" }, { "literal": "Bjoern Hartmann" }, { "literal": "Yang Li" } ], "citation-key": "arxiv_2407_08850", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "390a7415ad5f665de43f7d294e0a575785ef7e2810dead3363632174b2986af4", "submitted_at": "2024-07-11T20:18:19Z", "updated_at": "2024-08-13T23:41:43Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2407.08850v3" }, "canonical_source_url": "https://arxiv.org/abs/2407.08850", "catalog_id": "arxiv:2407.08850", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-07-11", "identifiers": { "arxiv": "2407.08850" }, "record_updated_at": "2024-08-13" }, "genre": "preprint", "id": "arxiv:2407.08850", "issued": { "date-parts": [ [ 2024, 7, 11 ] ] }, "note": "First published: 2024-07-11; Recorded arXiv version: v3; Recorded version updated: 2024-08-13T23:41:43Z.", "title": "UICrit: Enhancing Automated Design Evaluation with a UICritique Dataset", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2407.09018v1", "archive": "arXiv", "archive_location": "2407.09018", "author": [ { "literal": "Yongxiang Hu" }, { "literal": "Xuan Wang" }, { "literal": "Yingchuan Wang" }, { "literal": "Yu Zhang" }, { "literal": "Shiyu Guo" }, { "literal": "Chaoyi Chen" }, { "literal": "Xin Wang" }, { "literal": "Yangfan Zhou" } ], "citation-key": "arxiv_2407_09018", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a594fa7e150c9f5c22ed4d88f0551305f801f5780851ecf844802f2d2cd07c48", "submitted_at": "2024-07-12T06:14:46Z", "updated_at": "2024-07-12T06:14:46Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2407.09018v1" }, "canonical_source_url": "https://arxiv.org/abs/2407.09018", "catalog_id": "arxiv:2407.09018", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-07-12", "identifiers": { "arxiv": "2407.09018" }, "record_updated_at": "2024-07-12" }, "genre": "preprint", "id": "arxiv:2407.09018", "issued": { "date-parts": [ [ 2024, 7, 12 ] ] }, "note": "First published: 2024-07-12; Recorded arXiv version: v1; Recorded version updated: 2024-07-12T06:14:46Z.", "title": "AUITestAgent: Automatic Requirements Oriented GUI Function Testing", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2407.09295v3", "archive": "arXiv", "archive_location": "2407.09295", "author": [ { "literal": "Yulong Yang" }, { "literal": "Xinshan Yang" }, { "literal": "Shuaidong Li" }, { "literal": "Chenhao Lin" }, { "literal": "Zhengyu Zhao" }, { "literal": "Chao Shen" }, { "literal": "Tianwei Zhang" } ], "citation-key": "arxiv_2407_09295", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "88576325385d3c361f0666311eae8858ea0d4a0098d73221f789550bebf1e5db", "submitted_at": "2024-07-12T14:30:05Z", "updated_at": "2025-03-16T07:13:53Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2407.09295v3" }, "canonical_source_url": "https://arxiv.org/abs/2407.09295", "catalog_id": "arxiv:2407.09295", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-07-12", "identifiers": { "arxiv": "2407.09295" }, "record_updated_at": "2025-03-16" }, "genre": "preprint", "id": "arxiv:2407.09295", "issued": { "date-parts": [ [ 2024, 7, 12 ] ] }, "note": "First published: 2024-07-12; Recorded arXiv version: v3; Recorded version updated: 2025-03-16T07:13:53Z.", "title": "Systematic Categorization, Construction and Evaluation of New Attacks against Multi-modal Mobile GUI Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2407.10956v1", "archive": "arXiv", "archive_location": "2407.10956", "author": [ { "literal": "Ruisheng Cao" }, { "literal": "Fangyu Lei" }, { "literal": "Haoyuan Wu" }, { "literal": "Jixuan Chen" }, { "literal": "Yeqiao Fu" }, { "literal": "Hongcheng Gao" }, { "literal": "Xinzhuang Xiong" }, { "literal": "Hanchong Zhang" }, { "literal": "Yuchen Mao" }, { "literal": "Wenjing Hu" }, { "literal": "Tianbao Xie" }, { "literal": "Hongshen Xu" }, { "literal": "Danyang Zhang" }, { "literal": "Sida Wang" }, { "literal": "Ruoxi Sun" }, { "literal": "Pengcheng Yin" }, { "literal": "Caiming Xiong" }, { "literal": "Ansong Ni" }, { "literal": "Qian Liu" }, { "literal": "Victor Zhong" }, { "literal": "Lu Chen" }, { "literal": "Kai Yu" }, { "literal": "Tao Yu" } ], "citation-key": "arxiv_2407_10956", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7922a723142ac62018064fe5fc9357fb77c08e316ac4d6d6ba728f1b337df690", "submitted_at": "2024-07-15T17:54:37Z", "updated_at": "2024-07-15T17:54:37Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2407.10956v1" }, "canonical_source_url": "https://arxiv.org/abs/2407.10956", "catalog_id": "arxiv:2407.10956", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-07-15", "identifiers": { "arxiv": "2407.10956" }, "publication": { "authors": [ "Ruisheng Cao", "Fangyu Lei", "Haoyuan Wu", "Jixuan Chen", "Yeqiao Fu", "Hongcheng Gao", "Xinzhuang Xiong", "Hanchong Zhang", "Yuchen Mao", "Wenjing Hu", "Tianbao Xie", "Hongshen Xu", "Danyang Zhang", "Sida Wang", "Ruoxi Sun", "Pengcheng Yin", "Caiming Xiong", "Ansong Ni", "Qian Liu", "Victor Zhong", "Lu Chen", "Kai Yu", "Tao Yu" ], "date_basis": "Explicit publisher citation_publication_date: 2024-12-16. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 12, 16 ], "doi": "10.52202/079017-3421", "event_dates": null, "evidence_urls": [ "https://proceedings.neurips.cc/paper_files/paper/2024/hash/c2f71567cd53464161cab3336e8fc865-Abstract-Datasets_and_Benchmarks_Track.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity. The normalized publisher abstract matches the retained arXiv abstract SHA-256.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "107703-107744", "pdf_url": "https://proceedings.neurips.cc/paper_files/paper/2024/file/c2f71567cd53464161cab3336e8fc865-Paper-Datasets_and_Benchmarks_Track.pdf", "publisher": null, "series": null, "title": "Spider2-V: How Far Are Multimodal Agents From Automating Data Science and Engineering Workflows?", "type": "conference-paper", "url": "https://proceedings.neurips.cc/paper_files/paper/2024/hash/c2f71567cd53464161cab3336e8fc865-Abstract-Datasets_and_Benchmarks_Track.html", "venue": "Advances in Neural Information Processing Systems", "volume": "37" }, "record_updated_at": "2024-07-15" }, "genre": "preprint", "id": "arxiv:2407.10956", "issued": { "date-parts": [ [ 2024, 7, 15 ] ] }, "note": "First published: 2024-07-15; Recorded arXiv version: v1; Recorded version updated: 2024-07-15T17:54:37Z.", "title": "Spider2-V: How Far Are Multimodal Agents From Automating Data Science and Engineering Workflows?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2407.15711v2", "archive": "arXiv", "archive_location": "2407.15711", "author": [ { "literal": "Ori Yoran" }, { "literal": "Samuel Joseph Amouyal" }, { "literal": "Chaitanya Malaviya" }, { "literal": "Ben Bogin" }, { "literal": "Ofir Press" }, { "literal": "Jonathan Berant" } ], "citation-key": "arxiv_2407_15711", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2f028fb7f4d64fd4c29886d49b5d6f0fabddba309b669e019edf13586deb9c90", "submitted_at": "2024-07-22T15:18:45Z", "updated_at": "2024-10-21T15:45:31Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2407.15711v2" }, "canonical_source_url": "https://arxiv.org/abs/2407.15711", "catalog_id": "arxiv:2407.15711", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-07-22", "identifiers": { "arxiv": "2407.15711" }, "publication": { "authors": [ "Ori Yoran", "Samuel Joseph Amouyal", "Chaitanya Malaviya", "Ben Bogin", "Ofir Press", "Jonathan Berant" ], "date_basis": "Explicit publisher citation_publication_date: 2024/11. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2024, 11 ], "doi": "10.18653/v1/2024.emnlp-main.505", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2024.emnlp-main.505/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "8938-8968", "pdf_url": "https://aclanthology.org/2024.emnlp-main.505.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?", "type": "conference-paper", "url": "https://aclanthology.org/2024.emnlp-main.505/", "venue": "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing", "volume": null }, "record_updated_at": "2024-10-21" }, "genre": "preprint", "id": "arxiv:2407.15711", "issued": { "date-parts": [ [ 2024, 7, 22 ] ] }, "note": "First published: 2024-07-22; Recorded arXiv version: v2; Recorded version updated: 2024-10-21T15:45:31Z.", "title": "AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2407.18901v1", "archive": "arXiv", "archive_location": "2407.18901", "author": [ { "literal": "Harsh Trivedi" }, { "literal": "Tushar Khot" }, { "literal": "Mareike Hartmann" }, { "literal": "Ruskin Manku" }, { "literal": "Vinty Dong" }, { "literal": "Edward Li" }, { "literal": "Shashank Gupta" }, { "literal": "Ashish Sabharwal" }, { "literal": "Niranjan Balasubramanian" } ], "citation-key": "arxiv_2407_18901", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e59087ecf28d07baf81b60e4166a866d4bea4d8cd934bea06c2149dd61f38f1a", "submitted_at": "2024-07-26T17:55:45Z", "updated_at": "2024-07-26T17:55:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2407.18901v1" }, "canonical_source_url": "https://arxiv.org/abs/2407.18901", "catalog_id": "arxiv:2407.18901", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-07-26", "identifiers": { "arxiv": "2407.18901" }, "record_updated_at": "2024-07-26" }, "genre": "preprint", "id": "arxiv:2407.18901", "issued": { "date-parts": [ [ 2024, 7, 26 ] ] }, "note": "First published: 2024-07-26; Recorded arXiv version: v1; Recorded version updated: 2024-07-26T17:55:45Z.", "title": "AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2407.19056v1", "archive": "arXiv", "archive_location": "2407.19056", "author": [ { "literal": "Zilong Wang" }, { "literal": "Yuedong Cui" }, { "literal": "Li Zhong" }, { "literal": "Zimin Zhang" }, { "literal": "Da Yin" }, { "literal": "Bill Yuchen Lin" }, { "literal": "Jingbo Shang" } ], "citation-key": "arxiv_2407_19056", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c443e6a508c438332ac5990cf254a4441f5bdcc24354fd8aeca36bcbf4604a29", "submitted_at": "2024-07-26T19:27:17Z", "updated_at": "2024-07-26T19:27:17Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2407.19056v1" }, "canonical_source_url": "https://arxiv.org/abs/2407.19056", "catalog_id": "arxiv:2407.19056", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-07-26", "identifiers": { "arxiv": "2407.19056" }, "record_updated_at": "2024-07-26" }, "genre": "preprint", "id": "arxiv:2407.19056", "issued": { "date-parts": [ [ 2024, 7, 26 ] ] }, "note": "First published: 2024-07-26; Recorded arXiv version: v1; Recorded version updated: 2024-07-26T19:27:17Z.", "title": "OfficeBench: Benchmarking Language Agents across Multiple Applications for Office Automation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2408.02544v3", "archive": "arXiv", "archive_location": "2408.02544", "author": [ { "literal": "Xinbei Ma" }, { "literal": "Yiting Wang" }, { "literal": "Yao Yao" }, { "literal": "Tongxin Yuan" }, { "literal": "Aston Zhang" }, { "literal": "Zhuosheng Zhang" }, { "literal": "Hai Zhao" } ], "citation-key": "arxiv_2408_02544", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d66c0c2102dd2734d5334faeb12b4971d50a2d2e50a6671ede2fa78a84a0c03f", "submitted_at": "2024-08-05T15:16:22Z", "updated_at": "2025-09-05T09:21:10Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2408.02544v3" }, "canonical_source_url": "https://arxiv.org/abs/2408.02544", "catalog_id": "arxiv:2408.02544", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-08-05", "identifiers": { "arxiv": "2408.02544" }, "record_updated_at": "2025-09-05" }, "genre": "preprint", "id": "arxiv:2408.02544", "issued": { "date-parts": [ [ 2024, 8, 5 ] ] }, "note": "First published: 2024-08-05; Recorded arXiv version: v3; Recorded version updated: 2025-09-05T09:21:10Z.", "title": "Caution for the Environment: Multimodal LLM Agents are Susceptible to Environmental Distractions", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2408.06327v1", "archive": "arXiv", "archive_location": "2408.06327", "author": [ { "literal": "Xiao Liu" }, { "literal": "Tianjie Zhang" }, { "literal": "Yu Gu" }, { "literal": "Iat Long Iong" }, { "literal": "Yifan Xu" }, { "literal": "Xixuan Song" }, { "literal": "Shudan Zhang" }, { "literal": "Hanyu Lai" }, { "literal": "Xinyi Liu" }, { "literal": "Hanlin Zhao" }, { "literal": "Jiadai Sun" }, { "literal": "Xinyue Yang" }, { "literal": "Yu Yang" }, { "literal": "Zehan Qi" }, { "literal": "Shuntian Yao" }, { "literal": "Xueqiao Sun" }, { "literal": "Siyi Cheng" }, { "literal": "Qinkai Zheng" }, { "literal": "Hao Yu" }, { "literal": "Hanchen Zhang" }, { "literal": "Wenyi Hong" }, { "literal": "Ming Ding" }, { "literal": "Lihang Pan" }, { "literal": "Xiaotao Gu" }, { "literal": "Aohan Zeng" }, { "literal": "Zhengxiao Du" }, { "literal": "Chan Hee Song" }, { "literal": "Yu Su" }, { "literal": "Yuxiao Dong" }, { "literal": "Jie Tang" } ], "citation-key": "arxiv_2408_06327", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "603bfa4357617bfeb7a718065e77c71d77614f7240258684938dd4e6fd1a20df", "submitted_at": "2024-08-12T17:44:17Z", "updated_at": "2024-08-12T17:44:17Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2408.06327v1" }, "canonical_source_url": "https://arxiv.org/abs/2408.06327", "catalog_id": "arxiv:2408.06327", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-08-12", "identifiers": { "arxiv": "2408.06327" }, "record_updated_at": "2024-08-12" }, "genre": "preprint", "id": "arxiv:2408.06327", "issued": { "date-parts": [ [ 2024, 8, 12 ] ] }, "note": "First published: 2024-08-12; Recorded arXiv version: v1; Recorded version updated: 2024-08-12T17:44:17Z.", "title": "VisualAgentBench: Towards Large Multimodal Models as Visual Foundation Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2409.01927v1", "archive": "arXiv", "archive_location": "2409.01927", "author": [ { "literal": "Segev Shlomov" }, { "literal": "Ben wiesel" }, { "literal": "Aviad Sela" }, { "literal": "Ido Levy" }, { "literal": "Liane Galanti" }, { "literal": "Roy Abitbol" } ], "citation-key": "arxiv_2409_01927", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cdb228e652218d7f8981d9d16ca29b465004675c510d1e9c5c7b543bec9dc454", "submitted_at": "2024-09-03T14:17:09Z", "updated_at": "2024-09-03T14:17:09Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2409.01927v1" }, "canonical_source_url": "https://arxiv.org/abs/2409.01927", "catalog_id": "arxiv:2409.01927", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-09-03", "identifiers": { "arxiv": "2409.01927" }, "record_updated_at": "2024-09-03" }, "genre": "preprint", "id": "arxiv:2409.01927", "issued": { "date-parts": [ [ 2024, 9, 3 ] ] }, "note": "First published: 2024-09-03; Recorded arXiv version: v1; Recorded version updated: 2024-09-03T14:17:09Z.", "title": "From Grounding to Planning: Benchmarking Bottlenecks in Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2409.08264v2", "archive": "arXiv", "archive_location": "2409.08264", "author": [ { "literal": "Rogerio Bonatti" }, { "literal": "Dan Zhao" }, { "literal": "Francesco Bonacci" }, { "literal": "Dillon Dupont" }, { "literal": "Sara Abdali" }, { "literal": "Yinheng Li" }, { "literal": "Yadong Lu" }, { "literal": "Justin Wagle" }, { "literal": "Kazuhito Koishida" }, { "literal": "Arthur Bucker" }, { "literal": "Lawrence Jang" }, { "literal": "Zack Hui" } ], "citation-key": "arxiv_2409_08264", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fe77f7f8ffd7490192ab54b7f6d49c0cb5b10be6df0a645bfb181b3be724118a", "submitted_at": "2024-09-12T17:56:43Z", "updated_at": "2024-09-13T20:17:13Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2409.08264v2" }, "canonical_source_url": "https://arxiv.org/abs/2409.08264", "catalog_id": "arxiv:2409.08264", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-09-12", "identifiers": { "arxiv": "2409.08264" }, "record_updated_at": "2024-09-13" }, "genre": "preprint", "id": "arxiv:2409.08264", "issued": { "date-parts": [ [ 2024, 9, 12 ] ] }, "note": "First published: 2024-09-12; Recorded arXiv version: v2; Recorded version updated: 2024-09-13T20:17:13Z.", "title": "Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2409.11295v5", "archive": "arXiv", "archive_location": "2409.11295", "author": [ { "literal": "Zeyi Liao" }, { "literal": "Lingbo Mo" }, { "literal": "Chejian Xu" }, { "literal": "Mintong Kang" }, { "literal": "Jiawei Zhang" }, { "literal": "Chaowei Xiao" }, { "literal": "Yuan Tian" }, { "literal": "Bo Li" }, { "literal": "Huan Sun" } ], "citation-key": "arxiv_2409_11295", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "897cdbb6aa52478b260ecba830e0a13527c2614454f7c766c9f5c8c9bdbfd4e3", "submitted_at": "2024-09-17T15:49:44Z", "updated_at": "2025-03-12T20:54:00Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2409.11295v5" }, "canonical_source_url": "https://arxiv.org/abs/2409.11295", "catalog_id": "arxiv:2409.11295", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-09-17", "identifiers": { "arxiv": "2409.11295" }, "record_updated_at": "2025-03-12" }, "genre": "preprint", "id": "arxiv:2409.11295", "issued": { "date-parts": [ [ 2024, 9, 17 ] ] }, "note": "First published: 2024-09-17; Recorded arXiv version: v5; Recorded version updated: 2025-03-12T20:54:00Z.", "title": "EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2409.12089v3", "archive": "arXiv", "archive_location": "2409.12089", "author": [ { "literal": "Wayne Chi" }, { "literal": "Ameet Talwalkar" }, { "literal": "Chris Donahue" } ], "citation-key": "arxiv_2409_12089", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a2f59976cec7adb9f9c88bd79bdfbeed040e4874b6d9ab0c229136f98101aa98", "submitted_at": "2024-09-18T16:04:10Z", "updated_at": "2024-10-06T21:25:34Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2409.12089v3" }, "canonical_source_url": "https://arxiv.org/abs/2409.12089", "catalog_id": "arxiv:2409.12089", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-09-18", "identifiers": { "arxiv": "2409.12089" }, "record_updated_at": "2024-10-06" }, "genre": "preprint", "id": "arxiv:2409.12089", "issued": { "date-parts": [ [ 2024, 9, 18 ] ] }, "note": "First published: 2024-09-18; Recorded arXiv version: v3; Recorded version updated: 2024-10-06T21:25:34Z.", "title": "The Impact of Element Ordering on LM Agent Performance", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2409.13711v2", "archive": "arXiv", "archive_location": "2409.13711", "author": [ { "literal": "Maria Wang" }, { "literal": "Srinivas Sunkara" }, { "literal": "Gilles Baechler" }, { "literal": "Jason Lin" }, { "literal": "Yun Zhu" }, { "literal": "Fedir Zubach" }, { "literal": "Lei Shu" }, { "literal": "Jindong Chen" } ], "citation-key": "arxiv_2409_13711", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1ca44e6c98c09a2261b2cf8f3b6d18584b2d2471934b2dcac8d5b72c17b32e61", "submitted_at": "2024-09-06T18:44:25Z", "updated_at": "2024-09-24T18:38:02Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2409.13711v2" }, "canonical_source_url": "https://arxiv.org/abs/2409.13711", "catalog_id": "arxiv:2409.13711", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-09-06", "identifiers": { "arxiv": "2409.13711" }, "record_updated_at": "2024-09-24" }, "genre": "preprint", "id": "arxiv:2409.13711", "issued": { "date-parts": [ [ 2024, 9, 6 ] ] }, "note": "First published: 2024-09-06; Recorded arXiv version: v2; Recorded version updated: 2024-09-24T18:38:02Z.", "title": "WebQuest: A Benchmark for Multimodal QA on Web Page Sequences", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2409.18060v2", "archive": "arXiv", "archive_location": "2409.18060", "author": [ { "literal": "Sabrina Haque" }, { "literal": "Christoph Csallner" } ], "citation-key": "arxiv_2409_18060", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2d927b7644f0c6b3a49fbdf1aa976a00f2409adf9a72e1c04333438f66f1014b", "submitted_at": "2024-09-26T17:01:33Z", "updated_at": "2024-10-07T19:04:38Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2409.18060v2" }, "canonical_source_url": "https://arxiv.org/abs/2409.18060", "catalog_id": "arxiv:2409.18060", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2024-09-26", "identifiers": { "arxiv": "2409.18060" }, "record_updated_at": "2024-10-07" }, "genre": "preprint", "id": "arxiv:2409.18060", "issued": { "date-parts": [ [ 2024, 9, 26 ] ] }, "note": "First published: 2024-09-26; Recorded arXiv version: v2; Recorded version updated: 2024-10-07T19:04:38Z.", "title": "Inferring Alt-text For UI Icons With Large Language Models During App Development", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2410.00689v2", "archive": "arXiv", "archive_location": "2410.00689", "author": [ { "literal": "Ruhana Azam" }, { "literal": "Tamer Abuelsaad" }, { "literal": "Aditya Vempaty" }, { "literal": "Ashish Jagmohan" } ], "citation-key": "arxiv_2410_00689", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "69febc7aabfe38a6da439f1b4f673e0c36bc34c6afd1e8b06481433c1f43b1aa", "submitted_at": "2024-10-01T13:43:55Z", "updated_at": "2024-10-11T15:42:52Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2410.00689v2" }, "canonical_source_url": "https://arxiv.org/abs/2410.00689", "catalog_id": "arxiv:2410.00689", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-10-01", "identifiers": { "arxiv": "2410.00689" }, "record_updated_at": "2024-10-11" }, "genre": "preprint", "id": "arxiv:2410.00689", "issued": { "date-parts": [ [ 2024, 10, 1 ] ] }, "note": "First published: 2024-10-01; Recorded arXiv version: v2; Recorded version updated: 2024-10-11T15:42:52Z.", "title": "Multimodal Auto Validation For Self-Refinement in Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2410.06703v7", "archive": "arXiv", "archive_location": "2410.06703", "author": [ { "literal": "Ido Levy" }, { "literal": "Ben Wiesel" }, { "literal": "Sami Marreed" }, { "literal": "Alon Oved" }, { "literal": "Avi Yaeli" }, { "literal": "Nir Mashkif" }, { "literal": "Segev Shlomov" } ], "citation-key": "arxiv_2410_06703", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ba2b599c2b3f09974ee41d329545b69773a3b408b6db0124c5b1073d5fe8f1cf", "submitted_at": "2024-10-09T09:13:38Z", "updated_at": "2026-06-04T09:59:28Z", "version": 7, "versioned_url": "https://arxiv.org/abs/2410.06703v7" }, "canonical_source_url": "https://arxiv.org/abs/2410.06703", "catalog_id": "arxiv:2410.06703", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-10-09", "identifiers": { "arxiv": "2410.06703" }, "record_updated_at": "2026-06-04" }, "genre": "preprint", "id": "arxiv:2410.06703", "issued": { "date-parts": [ [ 2024, 10, 9 ] ] }, "note": "First published: 2024-10-09; Recorded arXiv version: v7; Recorded version updated: 2026-06-04T09:59:28Z.", "title": "ST-WebAgentBench: A Benchmark for Evaluating Safety and Trustworthiness in Web Agents", "type": "article", "version": "7" }, { "URL": "https://arxiv.org/abs/2410.09006v2", "archive": "arXiv", "archive_location": "2410.09006", "author": [ { "literal": "Zhuohao Jerry Zhang" }, { "literal": "Eldon Schoop" }, { "literal": "Jeffrey Nichols" }, { "literal": "Anuj Mahajan" }, { "literal": "Amanda Swearngin" } ], "citation-key": "arxiv_2410_09006", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ff649a0792864bd40b7bf2f825f6a8a815eb60cc795ed4dc0c6b55a942477758", "submitted_at": "2024-10-11T17:24:00Z", "updated_at": "2025-03-22T18:01:37Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2410.09006v2" }, "canonical_source_url": "https://arxiv.org/abs/2410.09006", "catalog_id": "arxiv:2410.09006", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-10-11", "identifiers": { "arxiv": "2410.09006" }, "record_updated_at": "2025-03-22" }, "genre": "preprint", "id": "arxiv:2410.09006", "issued": { "date-parts": [ [ 2024, 10, 11 ] ] }, "note": "First published: 2024-10-11; Recorded arXiv version: v2; Recorded version updated: 2025-03-22T18:01:37Z.", "title": "From Interaction to Impact: Towards Safer AI Agents Through Understanding and Evaluating Mobile UI Operation Impacts", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2410.13886v2", "archive": "arXiv", "archive_location": "2410.13886", "author": [ { "literal": "Priyanshu Kumar" }, { "literal": "Elaine Lau" }, { "literal": "Saranya Vijayakumar" }, { "literal": "Tu Trinh" }, { "literal": "Scale Red Team" }, { "literal": "Elaine Chang" }, { "literal": "Vaughn Robinson" }, { "literal": "Sean Hendryx" }, { "literal": "Shuyan Zhou" }, { "literal": "Matt Fredrikson" }, { "literal": "Summer Yue" }, { "literal": "Zifan Wang" } ], "citation-key": "arxiv_2410_13886", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4922d9362b23bc0e41d47dacf590c6e9bd6e88cfa6418cf3517814131295270c", "submitted_at": "2024-10-11T06:54:12Z", "updated_at": "2024-10-21T18:06:58Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2410.13886v2" }, "canonical_source_url": "https://arxiv.org/abs/2410.13886", "catalog_id": "arxiv:2410.13886", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-10-11", "identifiers": { "arxiv": "2410.13886" }, "record_updated_at": "2024-10-21" }, "genre": "preprint", "id": "arxiv:2410.13886", "issued": { "date-parts": [ [ 2024, 10, 11 ] ] }, "note": "First published: 2024-10-11; Recorded arXiv version: v2; Recorded version updated: 2024-10-21T18:06:58Z.", "title": "Refusal-Trained LLMs Are Easily Jailbroken As Browser Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2410.15164v3", "archive": "arXiv", "archive_location": "2410.15164", "author": [ { "literal": "Jingxuan Chen" }, { "literal": "Derek Yuen" }, { "literal": "Bin Xie" }, { "literal": "Yuhao Yang" }, { "literal": "Gongwei Chen" }, { "literal": "Zhihao Wu" }, { "literal": "Li Yixing" }, { "literal": "Xurui Zhou" }, { "literal": "Weiwen Liu" }, { "literal": "Shuai Wang" }, { "literal": "Kaiwen Zhou" }, { "literal": "Rui Shao" }, { "literal": "Liqiang Nie" }, { "literal": "Yasheng Wang" }, { "literal": "Jianye Hao" }, { "literal": "Jun Wang" }, { "literal": "Kun Shao" } ], "citation-key": "arxiv_2410_15164", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "206059754652205f1e036fac8ee117f71dd356bfe2d35fd84aff8d2935ea242d", "submitted_at": "2024-10-19T17:28:48Z", "updated_at": "2025-03-31T20:39:17Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2410.15164v3" }, "canonical_source_url": "https://arxiv.org/abs/2410.15164", "catalog_id": "arxiv:2410.15164", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-10-19", "identifiers": { "arxiv": "2410.15164" }, "publication": { "authors": [ "Jingxuan Chen", "Derek Yuen", "Bin Xie", "Yuhao Yang", "Gongwei Chen", "Zhihao Wu", "Li Yixing", "Xurui Zhou", "Weiwen Liu", "Shuai Wang", "Kaiwen Zhou", "Rui Shao", "Liqiang Nie", "Yasheng Wang", "Jianye HAO", "Jun Wang", "Kun Shao" ], "date_basis": "Explicit publisher citation_publication_date: 2025-05-01. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 5, 1 ], "doi": null, "event_dates": null, "evidence_urls": [ "https://proceedings.iclr.cc/paper_files/paper/2025/hash/9a75f4dd9679aa4ff80a4e6f0a1dc700-Abstract-Conference.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "61769-61806", "pdf_url": "https://proceedings.iclr.cc/paper_files/paper/2025/file/9a75f4dd9679aa4ff80a4e6f0a1dc700-Paper-Conference.pdf", "publisher": null, "series": null, "title": "SPA-BENCH: A COMPREHENSIVE BENCHMARK FOR SMARTPHONE AGENT EVALUATION", "type": "conference-paper", "url": "https://proceedings.iclr.cc/paper_files/paper/2025/hash/9a75f4dd9679aa4ff80a4e6f0a1dc700-Abstract-Conference.html", "venue": "International Conference on Learning Representations", "volume": "2025" }, "record_updated_at": "2025-03-31" }, "genre": "preprint", "id": "arxiv:2410.15164", "issued": { "date-parts": [ [ 2024, 10, 19 ] ] }, "note": "First published: 2024-10-19; Recorded arXiv version: v3; Recorded version updated: 2025-03-31T20:39:17Z.", "title": "SPA-Bench: A Comprehensive Benchmark for SmartPhone Agent Evaluation", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2410.16464v3", "archive": "arXiv", "archive_location": "2410.16464", "author": [ { "literal": "Yueqi Song" }, { "literal": "Frank Xu" }, { "literal": "Shuyan Zhou" }, { "literal": "Graham Neubig" } ], "citation-key": "arxiv_2410_16464", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0944387dd3c65ea7b2f5ca52721496cf10b3c9860b303063201e7811318f4950", "submitted_at": "2024-10-21T19:46:06Z", "updated_at": "2025-06-16T22:23:37Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2410.16464v3" }, "canonical_source_url": "https://arxiv.org/abs/2410.16464", "catalog_id": "arxiv:2410.16464", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-10-21", "identifiers": { "arxiv": "2410.16464" }, "publication": { "authors": [ "Yueqi Song", "Frank F. Xu", "Shuyan Zhou", "Graham Neubig" ], "date_basis": "Explicit publisher citation_publication_date: 2025/7. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 7 ], "doi": "10.18653/v1/2025.findings-acl.577", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.findings-acl.577/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "11066-11085", "pdf_url": "https://aclanthology.org/2025.findings-acl.577.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "Beyond Browsing: API-Based Web Agents", "type": "conference-paper", "url": "https://aclanthology.org/2025.findings-acl.577/", "venue": "Findings of the Association for Computational Linguistics: ACL 2025", "volume": null }, "record_updated_at": "2025-06-16" }, "genre": "preprint", "id": "arxiv:2410.16464", "issued": { "date-parts": [ [ 2024, 10, 21 ] ] }, "note": "First published: 2024-10-21; Recorded arXiv version: v3; Recorded version updated: 2025-06-16T22:23:37Z.", "title": "Beyond Browsing: API-Based Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2410.17401v4", "archive": "arXiv", "archive_location": "2410.17401", "author": [ { "literal": "Chejian Xu" }, { "literal": "Mintong Kang" }, { "literal": "Jiawei Zhang" }, { "literal": "Zeyi Liao" }, { "literal": "Lingbo Mo" }, { "literal": "Mengqi Yuan" }, { "literal": "Huan Sun" }, { "literal": "Bo Li" } ], "citation-key": "arxiv_2410_17401", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "226a4acbbaab42c8cbc44a897651d8c745119426ea1223f1b849558af32fab02", "submitted_at": "2024-10-22T20:18:26Z", "updated_at": "2025-05-31T18:34:01Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2410.17401v4" }, "canonical_source_url": "https://arxiv.org/abs/2410.17401", "catalog_id": "arxiv:2410.17401", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-10-22", "identifiers": { "arxiv": "2410.17401" }, "record_updated_at": "2025-05-31" }, "genre": "preprint", "id": "arxiv:2410.17401", "issued": { "date-parts": [ [ 2024, 10, 22 ] ] }, "note": "First published: 2024-10-22; Recorded arXiv version: v4; Recorded version updated: 2025-05-31T18:34:01Z.", "title": "AdvAgent: Controllable Blackbox Red-teaming on Web Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2410.17520v3", "archive": "arXiv", "archive_location": "2410.17520", "author": [ { "literal": "Juyong Lee" }, { "literal": "Dongyoon Hahm" }, { "literal": "June Suk Choi" }, { "literal": "W. Bradley Knox" }, { "literal": "Kimin Lee" } ], "citation-key": "arxiv_2410_17520", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "362cbf7ea267ddc4a64505aa95c4f16a9f3631f95319ce10eef9981e2c85937b", "submitted_at": "2024-10-23T02:51:43Z", "updated_at": "2026-01-27T01:51:55Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2410.17520v3" }, "canonical_source_url": "https://arxiv.org/abs/2410.17520", "catalog_id": "arxiv:2410.17520", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-10-23", "identifiers": { "arxiv": "2410.17520" }, "record_updated_at": "2026-01-27" }, "genre": "preprint", "id": "arxiv:2410.17520", "issued": { "date-parts": [ [ 2024, 10, 23 ] ] }, "note": "First published: 2024-10-23; Recorded arXiv version: v3; Recorded version updated: 2026-01-27T01:51:55Z.", "title": "MobileSafetyBench: Evaluating Safety of Autonomous Agents in Mobile Device Control", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2410.19100v3", "archive": "arXiv", "archive_location": "2410.19100", "author": [ { "literal": "Lawrence Jang" }, { "literal": "Yinheng Li" }, { "literal": "Dan Zhao" }, { "literal": "Charles Ding" }, { "literal": "Justin Lin" }, { "literal": "Paul Pu Liang" }, { "literal": "Rogerio Bonatti" }, { "literal": "Kazuhito Koishida" } ], "citation-key": "arxiv_2410_19100", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6f413943bd5c2a395ddb5cf38a1a5ad61754b35a9eb5de43e8d980bd83e4adff", "submitted_at": "2024-10-24T19:03:01Z", "updated_at": "2025-02-15T05:19:38Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2410.19100v3" }, "canonical_source_url": "https://arxiv.org/abs/2410.19100", "catalog_id": "arxiv:2410.19100", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-10-24", "identifiers": { "arxiv": "2410.19100" }, "publication": { "authors": [ "Lawrence Jang", "Yinheng Li", "Dan Zhao", "Charles Ding", "Justin Lin", "Paul Pu Liang", "Rogerio Bonatti", "Kazuhito Koishida" ], "date_basis": "Explicit publisher citation_publication_date: 2025-05-01. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 5, 1 ], "doi": null, "event_dates": null, "evidence_urls": [ "https://proceedings.iclr.cc/paper_files/paper/2025/hash/5b555804d495321df2e3208cc27f4fbc-Abstract-Conference.html" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "36934-36958", "pdf_url": "https://proceedings.iclr.cc/paper_files/paper/2025/file/5b555804d495321df2e3208cc27f4fbc-Paper-Conference.pdf", "publisher": null, "series": null, "title": "VideoWebArena: Evaluating Long Context Multimodal Agents with Video Understanding Web Tasks", "type": "conference-paper", "url": "https://proceedings.iclr.cc/paper_files/paper/2025/hash/5b555804d495321df2e3208cc27f4fbc-Abstract-Conference.html", "venue": "International Conference on Learning Representations", "volume": "2025" }, "record_updated_at": "2025-02-15" }, "genre": "preprint", "id": "arxiv:2410.19100", "issued": { "date-parts": [ [ 2024, 10, 24 ] ] }, "note": "First published: 2024-10-24; Recorded arXiv version: v3; Recorded version updated: 2025-02-15T05:19:38Z.", "title": "VideoWebArena: Evaluating Long Context Multimodal Agents with Video Understanding Web Tasks", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2410.23252v3", "archive": "arXiv", "archive_location": "2410.23252", "author": [ { "literal": "Haoyi Qiu" }, { "literal": "Alexander R. Fabbri" }, { "literal": "Divyansh Agarwal" }, { "literal": "Kung-Hsiang Huang" }, { "literal": "Sarah Tan" }, { "literal": "Nanyun Peng" }, { "literal": "Chien-Sheng Wu" } ], "citation-key": "arxiv_2410_23252", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "26fac89c78ab03fcbaf25bbfe93ebce090dec9adbf5c6dd5440641f154cffac9", "submitted_at": "2024-10-30T17:35:44Z", "updated_at": "2025-03-08T23:37:49Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2410.23252v3" }, "canonical_source_url": "https://arxiv.org/abs/2410.23252", "catalog_id": "arxiv:2410.23252", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-10-30", "identifiers": { "arxiv": "2410.23252" }, "publication": { "authors": [ "Haoyi Qiu", "Alexander Richard Fabbri", "Divyansh Agarwal", "Kung-Hsiang Huang", "Sarah Tan", "Nanyun Peng", "Chien-Sheng Wu" ], "date_basis": "Explicit publisher citation_publication_date: 2025/4. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 4 ], "doi": "10.18653/v1/2025.findings-naacl.222", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.findings-naacl.222/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "3978-4005", "pdf_url": "https://aclanthology.org/2025.findings-naacl.222.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "Evaluating Cultural and Social Awareness of LLM Web Agents", "type": "conference-paper", "url": "https://aclanthology.org/2025.findings-naacl.222/", "venue": "Findings of the Association for Computational Linguistics: NAACL 2025", "volume": null }, "record_updated_at": "2025-03-08" }, "genre": "preprint", "id": "arxiv:2410.23252", "issued": { "date-parts": [ [ 2024, 10, 30 ] ] }, "note": "First published: 2024-10-30; Recorded arXiv version: v3; Recorded version updated: 2025-03-08T23:37:49Z.", "title": "Evaluating Cultural and Social Awareness of LLM Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2410.24024v2", "archive": "arXiv", "archive_location": "2410.24024", "author": [ { "literal": "Yifan Xu" }, { "literal": "Xiao Liu" }, { "literal": "Xueqiao Sun" }, { "literal": "Siyi Cheng" }, { "literal": "Hao Yu" }, { "literal": "Hanyu Lai" }, { "literal": "Shudan Zhang" }, { "literal": "Dan Zhang" }, { "literal": "Jie Tang" }, { "literal": "Yuxiao Dong" } ], "citation-key": "arxiv_2410_24024", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d29fe5262adf8677bbd5fead628c5fc28707af263aa7e800d1dee1f7e499784c", "submitted_at": "2024-10-31T15:25:20Z", "updated_at": "2024-11-04T05:57:31Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2410.24024v2" }, "canonical_source_url": "https://arxiv.org/abs/2410.24024", "catalog_id": "arxiv:2410.24024", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-10-31", "identifiers": { "arxiv": "2410.24024" }, "publication": { "authors": [ "Yifan Xu", "Xiao Liu", "Xueqiao Sun", "Siyi Cheng", "Hao Yu (余浩)", "Hanyu Lai", "Shudan Zhang", "Dan Zhang", "Jie Tang", "Yuxiao Dong" ], "date_basis": "Explicit publisher citation_publication_date: 2025/7. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 7 ], "doi": "10.18653/v1/2025.acl-long.107", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.acl-long.107/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot. Publisher spelling, initials, name annotations or author lists differ from the retained preprint.", "pages": "2144-2166", "pdf_url": "https://aclanthology.org/2025.acl-long.107.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "AndroidLab: Training and Systematic Benchmarking of Android Autonomous Agents", "type": "conference-paper", "url": "https://aclanthology.org/2025.acl-long.107/", "venue": "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", "volume": null }, "record_updated_at": "2024-11-04" }, "genre": "preprint", "id": "arxiv:2410.24024", "issued": { "date-parts": [ [ 2024, 10, 31 ] ] }, "note": "First published: 2024-10-31; Recorded arXiv version: v2; Recorded version updated: 2024-11-04T05:57:31Z.", "title": "AndroidLab: Training and Systematic Benchmarking of Android Autonomous Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2411.02006v3", "archive": "arXiv", "archive_location": "2411.02006", "author": [ { "literal": "Biao Wu" }, { "literal": "Yanda Li" }, { "literal": "Zhiwei Zhang" }, { "literal": "Yunchao Wei" }, { "literal": "Meng Fang" }, { "literal": "Ling Chen" } ], "citation-key": "arxiv_2411_02006", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "de860a29f6c54317586c2218688436898f1d7a8b61207b1d5f4a65b05b3a4e6e", "submitted_at": "2024-11-04T11:50:58Z", "updated_at": "2025-09-14T09:13:39Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2411.02006v3" }, "canonical_source_url": "https://arxiv.org/abs/2411.02006", "catalog_id": "arxiv:2411.02006", "catalog_kind": "survey", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-11-04", "identifiers": { "arxiv": "2411.02006" }, "record_updated_at": "2025-09-14" }, "genre": "preprint", "id": "arxiv:2411.02006", "issued": { "date-parts": [ [ 2024, 11, 4 ] ] }, "note": "First published: 2024-11-04; Recorded arXiv version: v3; Recorded version updated: 2025-09-14T09:13:39Z.", "title": "Foundations and Recent Trends in Multimodal Mobile Agents: A Survey", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2411.02391v2", "archive": "arXiv", "archive_location": "2411.02391", "author": [ { "literal": "Yanzhe Zhang" }, { "literal": "Tao Yu" }, { "literal": "Diyi Yang" } ], "citation-key": "arxiv_2411_02391", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "085c64595365cc6439c8c5fa8a479259a3d59b7e9740e0f9d43e48d00e1439c0", "submitted_at": "2024-11-04T18:56:42Z", "updated_at": "2025-05-24T16:15:46Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2411.02391v2" }, "canonical_source_url": "https://arxiv.org/abs/2411.02391", "catalog_id": "arxiv:2411.02391", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2024-11-04", "identifiers": { "arxiv": "2411.02391" }, "publication": { "authors": [ "Yanzhe Zhang", "Tao Yu", "Diyi Yang" ], "date_basis": "Explicit publisher citation_publication_date: 2025/7. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 7 ], "doi": "10.18653/v1/2025.acl-long.411", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.acl-long.411/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "8387-8401", "pdf_url": "https://aclanthology.org/2025.acl-long.411.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "Attacking Vision-Language Computer Agents via Pop-ups", "type": "conference-paper", "url": "https://aclanthology.org/2025.acl-long.411/", "venue": "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", "volume": null }, "record_updated_at": "2025-05-24" }, "genre": "preprint", "id": "arxiv:2411.02391", "issued": { "date-parts": [ [ 2024, 11, 4 ] ] }, "note": "First published: 2024-11-04; Recorded arXiv version: v2; Recorded version updated: 2025-05-24T16:15:46Z.", "title": "Attacking Vision-Language Computer Agents via Pop-ups", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2411.10323v1", "archive": "arXiv", "archive_location": "2411.10323", "author": [ { "literal": "Siyuan Hu" }, { "literal": "Mingyu Ouyang" }, { "literal": "Difei Gao" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2411_10323", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "255e101e263c4e3a0cb491d4c34b3e75a8f3660cec2141e3a507a30cf5514dbd", "submitted_at": "2024-11-15T16:23:52Z", "updated_at": "2024-11-15T16:23:52Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2411.10323v1" }, "canonical_source_url": "https://arxiv.org/abs/2411.10323", "catalog_id": "arxiv:2411.10323", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-11-15", "identifiers": { "arxiv": "2411.10323" }, "record_updated_at": "2024-11-15" }, "genre": "preprint", "id": "arxiv:2411.10323", "issued": { "date-parts": [ [ 2024, 11, 15 ] ] }, "note": "First published: 2024-11-15; Recorded arXiv version: v1; Recorded version updated: 2024-11-15T16:23:52Z.", "title": "The Dawn of GUI Agent: A Preliminary Case Study with Claude 3.5 Computer Use", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2411.10943v1", "archive": "arXiv", "archive_location": "2411.10943", "author": [ { "literal": "Minghe Gao" }, { "literal": "Wendong Bu" }, { "literal": "Bingchen Miao" }, { "literal": "Yang Wu" }, { "literal": "Yunfei Li" }, { "literal": "Juncheng Li" }, { "literal": "Siliang Tang" }, { "literal": "Qi Wu" }, { "literal": "Yueting Zhuang" }, { "literal": "Meng Wang" } ], "citation-key": "arxiv_2411_10943", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "daa9b7092616e94a1d0124d36d8b75d0401b9647ec762c963524616a92def9a5", "submitted_at": "2024-11-17T02:44:56Z", "updated_at": "2024-11-17T02:44:56Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2411.10943v1" }, "canonical_source_url": "https://arxiv.org/abs/2411.10943", "catalog_id": "arxiv:2411.10943", "catalog_kind": "survey", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-11-17", "identifiers": { "arxiv": "2411.10943" }, "record_updated_at": "2024-11-17" }, "genre": "preprint", "id": "arxiv:2411.10943", "issued": { "date-parts": [ [ 2024, 11, 17 ] ] }, "note": "First published: 2024-11-17; Recorded arXiv version: v1; Recorded version updated: 2024-11-17T02:44:56Z.", "title": "Generalist Virtual Agents: A Survey on Autonomous Agents Across Digital Platforms", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2412.05467v4", "archive": "arXiv", "archive_location": "2412.05467", "author": [ { "literal": "Thibault Le Sellier De Chezelles" }, { "literal": "Maxime Gasse" }, { "literal": "Alexandre Drouin" }, { "literal": "Massimo Caccia" }, { "literal": "Léo Boisvert" }, { "literal": "Megh Thakkar" }, { "literal": "Tom Marty" }, { "literal": "Rim Assouel" }, { "literal": "Sahar Omidi Shayegan" }, { "literal": "Lawrence Keunho Jang" }, { "literal": "Xing Han Lù" }, { "literal": "Ori Yoran" }, { "literal": "Dehan Kong" }, { "literal": "Frank F. Xu" }, { "literal": "Siva Reddy" }, { "literal": "Quentin Cappart" }, { "literal": "Graham Neubig" }, { "literal": "Ruslan Salakhutdinov" }, { "literal": "Nicolas Chapados" }, { "literal": "Alexandre Lacoste" } ], "citation-key": "arxiv_2412_05467", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "572a993cd37a83bf7100dca71f0a44473ed6547f521409abf0f4ee22733a28cc", "submitted_at": "2024-12-06T23:43:59Z", "updated_at": "2025-02-28T16:02:27Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2412.05467v4" }, "canonical_source_url": "https://arxiv.org/abs/2412.05467", "catalog_id": "arxiv:2412.05467", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-12-06", "identifiers": { "arxiv": "2412.05467" }, "record_updated_at": "2025-02-28" }, "genre": "preprint", "id": "arxiv:2412.05467", "issued": { "date-parts": [ [ 2024, 12, 6 ] ] }, "note": "First published: 2024-12-06; Recorded arXiv version: v4; Recorded version updated: 2025-02-28T16:02:27Z.", "title": "The BrowserGym Ecosystem for Web Agent Research", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2412.10047v2", "archive": "arXiv", "archive_location": "2412.10047", "author": [ { "literal": "Lu Wang" }, { "literal": "Fangkai Yang" }, { "literal": "Chaoyun Zhang" }, { "literal": "Junting Lu" }, { "literal": "Jiaxu Qian" }, { "literal": "Shilin He" }, { "literal": "Pu Zhao" }, { "literal": "Bo Qiao" }, { "literal": "Ray Huang" }, { "literal": "Si Qin" }, { "literal": "Qisheng Su" }, { "literal": "Jiayi Ye" }, { "literal": "Yudi Zhang" }, { "literal": "Jian-Guang Lou" }, { "literal": "Qingwei Lin" }, { "literal": "Saravan Rajmohan" }, { "literal": "Dongmei Zhang" }, { "literal": "Qi Zhang" } ], "citation-key": "arxiv_2412_10047", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "54051d974e388ed2a26f1421b8c1fc79472bb941810519039ef48e90bf4761b6", "submitted_at": "2024-12-13T11:19:56Z", "updated_at": "2025-01-13T06:47:21Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2412.10047v2" }, "canonical_source_url": "https://arxiv.org/abs/2412.10047", "catalog_id": "arxiv:2412.10047", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-12-13", "identifiers": { "arxiv": "2412.10047" }, "record_updated_at": "2025-01-13" }, "genre": "preprint", "id": "arxiv:2412.10047", "issued": { "date-parts": [ [ 2024, 12, 13 ] ] }, "note": "First published: 2024-12-13; Recorded arXiv version: v2; Recorded version updated: 2025-01-13T06:47:21Z.", "title": "Large Action Models: From Inception to Implementation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2412.14161v3", "archive": "arXiv", "archive_location": "2412.14161", "author": [ { "literal": "Frank F. Xu" }, { "literal": "Yufan Song" }, { "literal": "Boxuan Li" }, { "literal": "Yuxuan Tang" }, { "literal": "Kritanjali Jain" }, { "literal": "Mengxue Bao" }, { "literal": "Zora Z. Wang" }, { "literal": "Xuhui Zhou" }, { "literal": "Zhitong Guo" }, { "literal": "Murong Cao" }, { "literal": "Mingyang Yang" }, { "literal": "Hao Yang Lu" }, { "literal": "Amaad Martin" }, { "literal": "Zhe Su" }, { "literal": "Leander Maben" }, { "literal": "Raj Mehta" }, { "literal": "Wayne Chi" }, { "literal": "Lawrence Jang" }, { "literal": "Yiqing Xie" }, { "literal": "Shuyan Zhou" }, { "literal": "Graham Neubig" } ], "citation-key": "arxiv_2412_14161", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "77f91dd9ec9f50fb5526a78cfe270a7a1705d5bd83a2aef26644453740c005d5", "submitted_at": "2024-12-18T18:55:40Z", "updated_at": "2025-09-10T08:35:19Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2412.14161v3" }, "canonical_source_url": "https://arxiv.org/abs/2412.14161", "catalog_id": "arxiv:2412.14161", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-12-18", "identifiers": { "arxiv": "2412.14161" }, "record_updated_at": "2025-09-10" }, "genre": "preprint", "id": "arxiv:2412.14161", "issued": { "date-parts": [ [ 2024, 12, 18 ] ] }, "note": "First published: 2024-12-18; Recorded arXiv version: v3; Recorded version updated: 2025-09-10T08:35:19Z.", "title": "TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2412.18426v1", "archive": "arXiv", "archive_location": "2412.18426", "author": [ { "literal": "Kangjia Zhao" }, { "literal": "Jiahui Song" }, { "literal": "Leigang Sha" }, { "literal": "Haozhan Shen" }, { "literal": "Zhi Chen" }, { "literal": "Tiancheng Zhao" }, { "literal": "Xiubo Liang" }, { "literal": "Jianwei Yin" } ], "citation-key": "arxiv_2412_18426", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9e509f015068bcf1294a97236ff3bbde5e3ceb167f3cb032fcfd84fc4a5b01d0", "submitted_at": "2024-12-24T13:41:47Z", "updated_at": "2024-12-24T13:41:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2412.18426v1" }, "canonical_source_url": "https://arxiv.org/abs/2412.18426", "catalog_id": "arxiv:2412.18426", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2024-12-24", "identifiers": { "arxiv": "2412.18426" }, "record_updated_at": "2024-12-24" }, "genre": "preprint", "id": "arxiv:2412.18426", "issued": { "date-parts": [ [ 2024, 12, 24 ] ] }, "note": "First published: 2024-12-24; Recorded arXiv version: v1; Recorded version updated: 2024-12-24T13:41:47Z.", "title": "GUI Testing Arena: A Unified Benchmark for Advancing Autonomous GUI Testing Agent", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2501.01149v3", "archive": "arXiv", "archive_location": "2501.01149", "author": [ { "literal": "Yuxiang Chai" }, { "literal": "Shunye Tang" }, { "literal": "Han Xiao" }, { "literal": "Weifeng Lin" }, { "literal": "Hanhao Li" }, { "literal": "Jiayu Zhang" }, { "literal": "Liang Liu" }, { "literal": "Pengxiang Zhao" }, { "literal": "Guangyi Liu" }, { "literal": "Guozhi Wang" }, { "literal": "Shuai Ren" }, { "literal": "Rongduo Han" }, { "literal": "Haining Zhang" }, { "literal": "Siyuan Huang" }, { "literal": "Hongsheng Li" } ], "citation-key": "arxiv_2501_01149", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8ac7100a0f3cc3f004672eb6923b1f0ecb43c3d98da6959f0ff7514d0fb2f442", "submitted_at": "2025-01-02T09:03:56Z", "updated_at": "2026-01-12T07:01:09Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2501.01149v3" }, "canonical_source_url": "https://arxiv.org/abs/2501.01149", "catalog_id": "arxiv:2501.01149", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-01-02", "identifiers": { "arxiv": "2501.01149" }, "record_updated_at": "2026-01-12" }, "genre": "preprint", "id": "arxiv:2501.01149", "issued": { "date-parts": [ [ 2025, 1, 2 ] ] }, "note": "First published: 2025-01-02; Recorded arXiv version: v3; Recorded version updated: 2026-01-12T07:01:09Z.", "title": "A3: Android Agent Arena for Mobile GUI Agents with Essential-State Procedural Evaluation", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2501.07572v3", "archive": "arXiv", "archive_location": "2501.07572", "author": [ { "literal": "Jialong Wu" }, { "literal": "Wenbiao Yin" }, { "literal": "Yong Jiang" }, { "literal": "Zhenglin Wang" }, { "literal": "Zekun Xi" }, { "literal": "Runnan Fang" }, { "literal": "Linhai Zhang" }, { "literal": "Yulan He" }, { "literal": "Deyu Zhou" }, { "literal": "Pengjun Xie" }, { "literal": "Fei Huang" } ], "citation-key": "arxiv_2501_07572", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "32cf550b1e1d5b671b3a4c834787cb67c32c01f508494c5610dac91e95bd92ba", "submitted_at": "2025-01-13T18:58:07Z", "updated_at": "2025-08-10T05:59:20Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2501.07572v3" }, "canonical_source_url": "https://arxiv.org/abs/2501.07572", "catalog_id": "arxiv:2501.07572", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-01-13", "identifiers": { "arxiv": "2501.07572" }, "record_updated_at": "2025-08-10" }, "genre": "preprint", "id": "arxiv:2501.07572", "issued": { "date-parts": [ [ 2025, 1, 13 ] ] }, "note": "First published: 2025-01-13; Recorded arXiv version: v3; Recorded version updated: 2025-08-10T05:59:20Z.", "title": "WebWalker: Benchmarking LLMs in Web Traversal", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2501.16609v4", "archive": "arXiv", "archive_location": "2501.16609", "author": [ { "literal": "Faria Huq" }, { "literal": "Zora Zhiruo Wang" }, { "literal": "Frank F. Xu" }, { "literal": "Tianyue Ou" }, { "literal": "Shuyan Zhou" }, { "literal": "Jeffrey P. Bigham" }, { "literal": "Graham Neubig" } ], "citation-key": "arxiv_2501_16609", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9b14b88273749ac54ebbe3694cbe4196640da8f81ba2faf77b0e6bbd79a060f6", "submitted_at": "2025-01-28T00:56:53Z", "updated_at": "2026-02-27T05:35:06Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2501.16609v4" }, "canonical_source_url": "https://arxiv.org/abs/2501.16609", "catalog_id": "arxiv:2501.16609", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2025-01-28", "identifiers": { "arxiv": "2501.16609" }, "publication": { "authors": [ "Faria Huq", "Zora Zhiruo Wang", "Frank F. Xu", "Tianyue Ou", "Shuyan Zhou", "Jeffrey P. Bigham", "Graham Neubig" ], "date_basis": "The ACL Anthology citation_publication_date is 2025/4, and its visible Month/Year fields state April 2025. Only month precision is supplied; no day is inferred from conference dates.", "date_parts": [ 2025, 4 ], "doi": "10.18653/v1/2025.naacl-demo.17", "event_dates": null, "evidence_urls": [ "https://arxiv.org/abs/2501.16609v4", "https://aclanthology.org/2025.naacl-demo.17/" ], "identity_basis": "The retained arXiv v4 abstract page explicitly links this DOI. The ACL publication has the same complete title and the same seven authors in the same order, and describes the same CowPilot human-agent navigation framework. This attaches the proceedings edition to the existing arXiv identity.", "issue": null, "notes": "The April 2025 proceedings edition is distinct from the retained arXiv v4, last revised on 2026-02-27. The publisher abstract has minor wording and typesetting differences from the retained arXiv abstract; neither its text nor its date replaces the original arXiv snapshot or annotation. Exact publication day, numeric volume, issue and series are not established.", "pages": "163-172", "pdf_url": "https://aclanthology.org/2025.naacl-demo.17.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "CowPilot: A Framework for Autonomous and Human-Agent Collaborative Web Navigation", "type": "conference-paper", "url": "https://aclanthology.org/2025.naacl-demo.17/", "venue": "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (System Demonstrations)", "volume": null }, "record_updated_at": "2026-02-27" }, "genre": "preprint", "id": "arxiv:2501.16609", "issued": { "date-parts": [ [ 2025, 1, 28 ] ] }, "note": "First published: 2025-01-28; Recorded arXiv version: v4; Recorded version updated: 2026-02-27T05:35:06Z.", "title": "CowPilot: A Framework for Autonomous and Human-Agent Collaborative Web Navigation", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2502.08047v5", "archive": "arXiv", "archive_location": "2502.08047", "author": [ { "literal": "Henry Hengyuan Zhao" }, { "literal": "Kaiming Yang" }, { "literal": "Wendi Yu" }, { "literal": "Difei Gao" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2502_08047", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "68e6ecfcdff0163debe96f6440fb33529356bf9fd2c8743b5953ae52ae71c9ef", "submitted_at": "2025-02-12T01:06:10Z", "updated_at": "2026-05-23T20:19:43Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2502.08047v5" }, "canonical_source_url": "https://arxiv.org/abs/2502.08047", "catalog_id": "arxiv:2502.08047", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-02-12", "identifiers": { "arxiv": "2502.08047" }, "record_updated_at": "2026-05-23" }, "genre": "preprint", "id": "arxiv:2502.08047", "issued": { "date-parts": [ [ 2025, 2, 12 ] ] }, "note": "First published: 2025-02-12; Recorded arXiv version: v5; Recorded version updated: 2026-05-23T20:19:43Z.", "title": "WorldGUI: An Interactive Benchmark for Desktop GUI Automation from Any Starting Point", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2502.13053v3", "archive": "arXiv", "archive_location": "2502.13053", "author": [ { "literal": "Yurun Chen" }, { "literal": "Xavier Hu" }, { "literal": "Keting Yin" }, { "literal": "Juncheng Li" }, { "literal": "Shengyu Zhang" } ], "citation-key": "arxiv_2502_13053", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5af8d3a8b4db150bab0630f6040a0f3bced612f68b11d31985053f1f25c0f6f6", "submitted_at": "2025-02-18T17:01:28Z", "updated_at": "2025-08-06T07:04:43Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2502.13053v3" }, "canonical_source_url": "https://arxiv.org/abs/2502.13053", "catalog_id": "arxiv:2502.13053", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-02-18", "identifiers": { "arxiv": "2502.13053" }, "record_updated_at": "2025-08-06" }, "genre": "preprint", "id": "arxiv:2502.13053", "issued": { "date-parts": [ [ 2025, 2, 18 ] ] }, "note": "First published: 2025-02-18; Recorded arXiv version: v3; Recorded version updated: 2025-08-06T07:04:43Z.", "title": "Evaluating the Robustness of Multimodal Agents Against Active Environmental Injection Attacks", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2502.17903v1", "archive": "arXiv", "archive_location": "2502.17903", "author": [ { "literal": "Lars Krupp" }, { "literal": "Daniel Geißler" }, { "literal": "Paul Lukowicz" }, { "literal": "Jakob Karolus" } ], "citation-key": "arxiv_2502_17903", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3903e97e6db6bd470699f0d7b2785e53e15291234f1e5f717bae8d2ccd851ca4", "submitted_at": "2025-02-25T06:58:40Z", "updated_at": "2025-02-25T06:58:40Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2502.17903v1" }, "canonical_source_url": "https://arxiv.org/abs/2502.17903", "catalog_id": "arxiv:2502.17903", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-02-25", "identifiers": { "arxiv": "2502.17903" }, "record_updated_at": "2025-02-25" }, "genre": "preprint", "id": "arxiv:2502.17903", "issued": { "date-parts": [ [ 2025, 2, 25 ] ] }, "note": "First published: 2025-02-25; Recorded arXiv version: v1; Recorded version updated: 2025-02-25T06:58:40Z.", "title": "Towards Sustainable Web Agents: A Plea for Transparency and Dedicated Metrics for Energy Consumption", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2502.18525v2", "archive": "arXiv", "archive_location": "2502.18525", "author": [ { "literal": "Pranjal Aggarwal" }, { "literal": "Sean Welleck" } ], "citation-key": "arxiv_2502_18525", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9e4091fc740249b6406dd9e96f2692d7fac9323759ec6fad6ba5a05f3e445844", "submitted_at": "2025-02-24T18:41:33Z", "updated_at": "2025-10-03T02:09:59Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2502.18525v2" }, "canonical_source_url": "https://arxiv.org/abs/2502.18525", "catalog_id": "arxiv:2502.18525", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-02-24", "identifiers": { "arxiv": "2502.18525" }, "record_updated_at": "2025-10-03" }, "genre": "preprint", "id": "arxiv:2502.18525", "issued": { "date-parts": [ [ 2025, 2, 24 ] ] }, "note": "First published: 2025-02-24; Recorded arXiv version: v2; Recorded version updated: 2025-10-03T02:09:59Z.", "title": "Programming with Pixels: Can Computer-Use Agents do Software Engineering?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2502.20383v3", "archive": "arXiv", "archive_location": "2502.20383", "author": [ { "literal": "Jeffrey Yang Fan Chiang" }, { "literal": "Seungjae Lee" }, { "literal": "Jia-Bin Huang" }, { "literal": "Furong Huang" }, { "literal": "Yizheng Chen" } ], "citation-key": "arxiv_2502_20383", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "27eacc2668411e2b7f79423940aa12293fa57aa81390c64587d3074c141dcb4f", "submitted_at": "2025-02-27T18:56:26Z", "updated_at": "2025-09-21T22:29:53Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2502.20383v3" }, "canonical_source_url": "https://arxiv.org/abs/2502.20383", "catalog_id": "arxiv:2502.20383", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-02-27", "identifiers": { "arxiv": "2502.20383" }, "record_updated_at": "2025-09-21" }, "genre": "preprint", "id": "arxiv:2502.20383", "issued": { "date-parts": [ [ 2025, 2, 27 ] ] }, "note": "First published: 2025-02-27; Recorded arXiv version: v3; Recorded version updated: 2025-09-21T22:29:53Z.", "title": "Why Are Web AI Agents More Vulnerable Than Standalone LLMs? A Security Analysis", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2503.01861v3", "archive": "arXiv", "archive_location": "2503.01861", "author": [ { "literal": "Sami Marreed" }, { "literal": "Alon Oved" }, { "literal": "Avi Yaeli" }, { "literal": "Segev Shlomov" }, { "literal": "Ido Levy" }, { "literal": "Offer Akrabi" }, { "literal": "Aviad Sela" }, { "literal": "Asaf Adi" }, { "literal": "Nir Mashkif" } ], "citation-key": "arxiv_2503_01861", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2248d423c954ee6e406e3f46f6da43fabd6e832ffacc05cfa32083f2ff0c2700", "submitted_at": "2025-02-24T09:31:56Z", "updated_at": "2025-07-09T13:52:32Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2503.01861v3" }, "canonical_source_url": "https://arxiv.org/abs/2503.01861", "catalog_id": "arxiv:2503.01861", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-02-24", "identifiers": { "arxiv": "2503.01861" }, "record_updated_at": "2025-07-09" }, "genre": "preprint", "id": "arxiv:2503.01861", "issued": { "date-parts": [ [ 2025, 2, 24 ] ] }, "note": "First published: 2025-02-24; Recorded arXiv version: v3; Recorded version updated: 2025-07-09T13:52:32Z.", "title": "Towards Enterprise-Ready Computer Using Generalist Agent", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2503.02403v2", "archive": "arXiv", "archive_location": "2503.02403", "author": [ { "literal": "Jiahui Sun" }, { "literal": "Zhichao Hua" }, { "literal": "Yubin Xia" } ], "citation-key": "arxiv_2503_02403", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a0c393d109ab0f8da2bf25bb39581236babd50f7554f2ec99679a99fb7215171", "submitted_at": "2025-03-04T08:44:30Z", "updated_at": "2025-09-24T05:11:34Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2503.02403v2" }, "canonical_source_url": "https://arxiv.org/abs/2503.02403", "catalog_id": "arxiv:2503.02403", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-04", "identifiers": { "arxiv": "2503.02403" }, "record_updated_at": "2025-09-24" }, "genre": "preprint", "id": "arxiv:2503.02403", "issued": { "date-parts": [ [ 2025, 3, 4 ] ] }, "note": "First published: 2025-03-04; Recorded arXiv version: v2; Recorded version updated: 2025-09-24T05:11:34Z.", "title": "AutoEval: A Practical Framework for Autonomous Evaluation of Mobile Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2503.04730v1", "archive": "arXiv", "archive_location": "2503.04730", "author": [ { "literal": "Zheng Hui" }, { "literal": "Yinheng Li" }, { "literal": "Dan zhao" }, { "literal": "Tianyi Chen" }, { "literal": "Colby Banbury" }, { "literal": "Kazuhito Koishida" } ], "citation-key": "arxiv_2503_04730", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cc499e361f0bc46c3c5ee5a8abebd911396e383a004655de56db1ec7cec8e231", "submitted_at": "2025-01-27T08:29:17Z", "updated_at": "2025-01-27T08:29:17Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2503.04730v1" }, "canonical_source_url": "https://arxiv.org/abs/2503.04730", "catalog_id": "arxiv:2503.04730", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-01-27", "identifiers": { "arxiv": "2503.04730" }, "record_updated_at": "2025-01-27" }, "genre": "preprint", "id": "arxiv:2503.04730", "issued": { "date-parts": [ [ 2025, 1, 27 ] ] }, "note": "First published: 2025-01-27; Recorded arXiv version: v1; Recorded version updated: 2025-01-27T08:29:17Z.", "title": "WinClick: GUI Grounding with Multimodal Large Language Models", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2503.04957v1", "archive": "arXiv", "archive_location": "2503.04957", "author": [ { "literal": "Ada Defne Tur" }, { "literal": "Nicholas Meade" }, { "literal": "Xing Han Lù" }, { "literal": "Alejandra Zambrano" }, { "literal": "Arkil Patel" }, { "literal": "Esin Durmus" }, { "literal": "Spandana Gella" }, { "literal": "Karolina Stańczak" }, { "literal": "Siva Reddy" } ], "citation-key": "arxiv_2503_04957", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e4588022dd47624058db8a16d7f7763c591241298e94594b56bfeb4ab2e01737", "submitted_at": "2025-03-06T20:43:14Z", "updated_at": "2025-03-06T20:43:14Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2503.04957v1" }, "canonical_source_url": "https://arxiv.org/abs/2503.04957", "catalog_id": "arxiv:2503.04957", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-06", "identifiers": { "arxiv": "2503.04957" }, "record_updated_at": "2025-03-06" }, "genre": "preprint", "id": "arxiv:2503.04957", "issued": { "date-parts": [ [ 2025, 3, 6 ] ] }, "note": "First published: 2025-03-06; Recorded arXiv version: v1; Recorded version updated: 2025-03-06T20:43:14Z.", "title": "SafeArena: Evaluating the Safety of Autonomous Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2503.05143v1", "archive": "arXiv", "archive_location": "2503.05143", "author": [ { "literal": "Wenhao Wang" }, { "literal": "Zijie Yu" }, { "literal": "Rui Ye" }, { "literal": "Jianqing Zhang" }, { "literal": "Siheng Chen" }, { "literal": "Yanfeng Wang" } ], "citation-key": "arxiv_2503_05143", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b408c850ae83e603a1b3d8303164b9f2d7be51580be99027275b0734ea8f56f2", "submitted_at": "2025-03-07T04:52:20Z", "updated_at": "2025-03-07T04:52:20Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2503.05143v1" }, "canonical_source_url": "https://arxiv.org/abs/2503.05143", "catalog_id": "arxiv:2503.05143", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-07", "identifiers": { "arxiv": "2503.05143" }, "record_updated_at": "2025-03-07" }, "genre": "preprint", "id": "arxiv:2503.05143", "issued": { "date-parts": [ [ 2025, 3, 7 ] ] }, "note": "First published: 2025-03-07; Recorded arXiv version: v1; Recorded version updated: 2025-03-07T04:52:20Z.", "title": "FedMABench: Benchmarking Mobile Agents on Decentralized Heterogeneous User Data", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2503.07919v3", "archive": "arXiv", "archive_location": "2503.07919", "author": [ { "literal": "Yixiao Song" }, { "literal": "Katherine Thai" }, { "literal": "Chau Minh Pham" }, { "literal": "Yapei Chang" }, { "literal": "Mazin Nadaf" }, { "literal": "Mohit Iyyer" } ], "citation-key": "arxiv_2503_07919", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e1cd6327559d11982fb4a97e739cb6b5234c509859692084ed78130241f6d252", "submitted_at": "2025-03-10T23:50:30Z", "updated_at": "2025-07-24T17:45:05Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2503.07919v3" }, "canonical_source_url": "https://arxiv.org/abs/2503.07919", "catalog_id": "arxiv:2503.07919", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-10", "identifiers": { "arxiv": "2503.07919" }, "record_updated_at": "2025-07-24" }, "genre": "preprint", "id": "arxiv:2503.07919", "issued": { "date-parts": [ [ 2025, 3, 10 ] ] }, "note": "First published: 2025-03-10; Recorded arXiv version: v3; Recorded version updated: 2025-07-24T17:45:05Z.", "title": "BEARCUBS: A benchmark for computer-using web agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2503.09241v1", "archive": "arXiv", "archive_location": "2503.09241", "author": [ { "literal": "Pei Yang" }, { "literal": "Hai Ci" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2503_09241", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7d4423ee1e7047a6228f5b29684618fc4108af0694484a7ceddf6c3b53fbb06c", "submitted_at": "2025-03-12T10:38:15Z", "updated_at": "2025-03-12T10:38:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2503.09241v1" }, "canonical_source_url": "https://arxiv.org/abs/2503.09241", "catalog_id": "arxiv:2503.09241", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-12", "identifiers": { "arxiv": "2503.09241" }, "record_updated_at": "2025-03-12" }, "genre": "preprint", "id": "arxiv:2503.09241", "issued": { "date-parts": [ [ 2025, 3, 12 ] ] }, "note": "First published: 2025-03-12; Recorded arXiv version: v1; Recorded version updated: 2025-03-12T10:38:15Z.", "title": "In-Context Defense in Computer Agents: An Empirical Study", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2503.09780v3", "archive": "arXiv", "archive_location": "2503.09780", "author": [ { "literal": "Arman Zharmagambetov" }, { "literal": "Chuan Guo" }, { "literal": "Ivan Evtimov" }, { "literal": "Maya Pavlova" }, { "literal": "Ruslan Salakhutdinov" }, { "literal": "Kamalika Chaudhuri" } ], "citation-key": "arxiv_2503_09780", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bbb0726446c9679781b8e5930da0d3f406989a4b47759ce3440c9ad56d5839d8", "submitted_at": "2025-03-12T19:30:31Z", "updated_at": "2025-10-01T23:55:58Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2503.09780v3" }, "canonical_source_url": "https://arxiv.org/abs/2503.09780", "catalog_id": "arxiv:2503.09780", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-12", "identifiers": { "arxiv": "2503.09780" }, "record_updated_at": "2025-10-01" }, "genre": "preprint", "id": "arxiv:2503.09780", "issued": { "date-parts": [ [ 2025, 3, 12 ] ] }, "note": "First published: 2025-03-12; Recorded arXiv version: v3; Recorded version updated: 2025-10-01T23:55:58Z.", "title": "AgentDAM: Privacy Leakage Evaluation for Autonomous Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2503.12532v2", "archive": "arXiv", "archive_location": "2503.12532", "author": [ { "literal": "Fanbin Lu" }, { "literal": "Zhisheng Zhong" }, { "literal": "Ziqin Wei" }, { "literal": "Shu Liu" }, { "literal": "Chi-Wing Fu" }, { "literal": "Jiaya Jia" } ], "citation-key": "arxiv_2503_12532", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e883289bafee01759a0f4f3f7051cbbebdcd8059e7520d68c25ba313b8da69a6", "submitted_at": "2025-03-16T14:53:43Z", "updated_at": "2025-03-24T16:33:28Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2503.12532v2" }, "canonical_source_url": "https://arxiv.org/abs/2503.12532", "catalog_id": "arxiv:2503.12532", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-16", "identifiers": { "arxiv": "2503.12532" }, "record_updated_at": "2025-03-24" }, "genre": "preprint", "id": "arxiv:2503.12532", "issued": { "date-parts": [ [ 2025, 3, 16 ] ] }, "note": "First published: 2025-03-16; Recorded arXiv version: v2; Recorded version updated: 2025-03-24T16:33:28Z.", "title": "STEVE: A Step Verification Pipeline for Computer-use Agent Training", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2503.15661v2", "archive": "arXiv", "archive_location": "2503.15661", "author": [ { "literal": "Shravan Nayak" }, { "literal": "Xiangru Jian" }, { "literal": "Kevin Qinghong Lin" }, { "literal": "Juan A. Rodriguez" }, { "literal": "Montek Kalsi" }, { "literal": "Rabiul Awal" }, { "literal": "Nicolas Chapados" }, { "literal": "M. Tamer Özsu" }, { "literal": "Aishwarya Agrawal" }, { "literal": "David Vazquez" }, { "literal": "Christopher Pal" }, { "literal": "Perouz Taslakian" }, { "literal": "Spandana Gella" }, { "literal": "Sai Rajeswar" } ], "citation-key": "arxiv_2503_15661", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5abf3c7b08a4331d4f634d8cc63a7f8ff40713f7344231887dac1b5e061a8eae", "submitted_at": "2025-03-19T19:26:17Z", "updated_at": "2025-05-06T17:43:11Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2503.15661v2" }, "canonical_source_url": "https://arxiv.org/abs/2503.15661", "catalog_id": "arxiv:2503.15661", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-19", "identifiers": { "arxiv": "2503.15661" }, "record_updated_at": "2025-05-06" }, "genre": "preprint", "id": "arxiv:2503.15661", "issued": { "date-parts": [ [ 2025, 3, 19 ] ] }, "note": "First published: 2025-03-19; Recorded arXiv version: v2; Recorded version updated: 2025-05-06T17:43:11Z.", "title": "UI-Vision: A Desktop-centric GUI Benchmark for Visual Perception and Interaction", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2503.16788v1", "archive": "arXiv", "archive_location": "2503.16788", "author": [ { "literal": "Li Zhang" }, { "literal": "Longxi Gao" }, { "literal": "Mengwei Xu" } ], "citation-key": "arxiv_2503_16788", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e450216f8f802d8f3d16c5886097cef801d051951af0c57fdf195cc176ac744f", "submitted_at": "2025-03-21T01:52:43Z", "updated_at": "2025-03-21T01:52:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2503.16788v1" }, "canonical_source_url": "https://arxiv.org/abs/2503.16788", "catalog_id": "arxiv:2503.16788", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-21", "identifiers": { "arxiv": "2503.16788" }, "record_updated_at": "2025-03-21" }, "genre": "preprint", "id": "arxiv:2503.16788", "issued": { "date-parts": [ [ 2025, 3, 21 ] ] }, "note": "First published: 2025-03-21; Recorded arXiv version: v1; Recorded version updated: 2025-03-21T01:52:43Z.", "title": "Does Chain-of-Thought Reasoning Help Mobile GUI Agent? An Empirical Study", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2503.19537v1", "archive": "arXiv", "archive_location": "2503.19537", "author": [ { "literal": "Noam Kahlon" }, { "literal": "Guy Rom" }, { "literal": "Anatoly Efros" }, { "literal": "Filippo Galgani" }, { "literal": "Omri Berkovitch" }, { "literal": "Sapir Caduri" }, { "literal": "William E. Bishop" }, { "literal": "Oriana Riva" }, { "literal": "Ido Dagan" } ], "citation-key": "arxiv_2503_19537", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8a49d662eddf84690395b025e50f3378c3be2a3fc8abe02bd7202c6ff879bbe3", "submitted_at": "2025-03-25T10:46:08Z", "updated_at": "2025-03-25T10:46:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2503.19537v1" }, "canonical_source_url": "https://arxiv.org/abs/2503.19537", "catalog_id": "arxiv:2503.19537", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-25", "identifiers": { "arxiv": "2503.19537" }, "record_updated_at": "2025-03-25" }, "genre": "preprint", "id": "arxiv:2503.19537", "issued": { "date-parts": [ [ 2025, 3, 25 ] ] }, "note": "First published: 2025-03-25; Recorded arXiv version: v1; Recorded version updated: 2025-03-25T10:46:08Z.", "title": "Agent-Initiated Interaction in Phone UI Automation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2503.20279v3", "archive": "arXiv", "archive_location": "2503.20279", "author": [ { "literal": "Sejin Lee" }, { "literal": "Jian Kim" }, { "literal": "Haon Park" }, { "literal": "Ashkan Yousefpour" }, { "literal": "Sangyoon Yu" }, { "literal": "Min Song" } ], "citation-key": "arxiv_2503_20279", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5e90cbc547703a2673cfb30bbe9079496ce4c076b51f39e0557f5db9048947c8", "submitted_at": "2025-03-26T07:08:15Z", "updated_at": "2025-06-08T05:39:56Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2503.20279v3" }, "canonical_source_url": "https://arxiv.org/abs/2503.20279", "catalog_id": "arxiv:2503.20279", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-26", "identifiers": { "arxiv": "2503.20279" }, "record_updated_at": "2025-06-08" }, "genre": "preprint", "id": "arxiv:2503.20279", "issued": { "date-parts": [ [ 2025, 3, 26 ] ] }, "note": "First published: 2025-03-26; Recorded arXiv version: v3; Recorded version updated: 2025-06-08T05:39:56Z.", "title": "sudo rm -rf agentic_security", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2503.24180v2", "archive": "arXiv", "archive_location": "2503.24180", "author": [ { "literal": "Ziming Cheng" }, { "literal": "Zhiyuan Huang" }, { "literal": "Junting Pan" }, { "literal": "Zhaohui Hou" }, { "literal": "Mingjie Zhan" } ], "citation-key": "arxiv_2503_24180", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1bb09b54379d2fbc5b489f24a27f6311b91bb0821eefe5187aa160ffb00639d9", "submitted_at": "2025-03-31T14:56:24Z", "updated_at": "2025-08-24T08:39:18Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2503.24180v2" }, "canonical_source_url": "https://arxiv.org/abs/2503.24180", "catalog_id": "arxiv:2503.24180", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-03-31", "identifiers": { "arxiv": "2503.24180" }, "record_updated_at": "2025-08-24" }, "genre": "preprint", "id": "arxiv:2503.24180", "issued": { "date-parts": [ [ 2025, 3, 31 ] ] }, "note": "First published: 2025-03-31; Recorded arXiv version: v2; Recorded version updated: 2025-08-24T08:39:18Z.", "title": "Navi-plus: Managing Ambiguous GUI Navigation Tasks with Follow-up Questions", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2504.01382v4", "archive": "arXiv", "archive_location": "2504.01382", "author": [ { "literal": "Tianci Xue" }, { "literal": "Weijian Qi" }, { "literal": "Tianneng Shi" }, { "literal": "Chan Hee Song" }, { "literal": "Boyu Gou" }, { "literal": "Dawn Song" }, { "literal": "Huan Sun" }, { "literal": "Yu Su" } ], "citation-key": "arxiv_2504_01382", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ccad667f7c5bb75e58d75d5eab5cc24c7c3610be9eed1a09079de5e23c85af04", "submitted_at": "2025-04-02T05:51:29Z", "updated_at": "2025-10-08T02:22:45Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2504.01382v4" }, "canonical_source_url": "https://arxiv.org/abs/2504.01382", "catalog_id": "arxiv:2504.01382", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-02", "identifiers": { "arxiv": "2504.01382" }, "record_updated_at": "2025-10-08" }, "genre": "preprint", "id": "arxiv:2504.01382", "issued": { "date-parts": [ [ 2025, 4, 2 ] ] }, "note": "First published: 2025-04-02; Recorded arXiv version: v4; Recorded version updated: 2025-10-08T02:22:45Z.", "title": "An Illusion of Progress? Assessing the Current State of Web Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2504.01495v1", "archive": "arXiv", "archive_location": "2504.01495", "author": [ { "literal": "Antoine Chevrot" }, { "literal": "Alexandre Vernotte" }, { "literal": "Jean-Rémy Falleri" }, { "literal": "Xavier Blanc" }, { "literal": "Bruno Legeard" } ], "citation-key": "arxiv_2504_01495", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "76d0ffb55478570499c8cddf92a27d346cf11749f8ae5f8e128c87653b2bdbdc", "submitted_at": "2025-04-02T08:48:01Z", "updated_at": "2025-04-02T08:48:01Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.01495v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.01495", "catalog_id": "arxiv:2504.01495", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-02", "identifiers": { "arxiv": "2504.01495" }, "record_updated_at": "2025-04-02" }, "genre": "preprint", "id": "arxiv:2504.01495", "issued": { "date-parts": [ [ 2025, 4, 2 ] ] }, "note": "First published: 2025-04-02; Recorded arXiv version: v1; Recorded version updated: 2025-04-02T08:48:01Z.", "title": "Are Autonomous Web Agents Good Testers?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.02357v1", "archive": "arXiv", "archive_location": "2504.02357", "author": [ { "literal": "Xiaolei Li" }, { "literal": "Jialun Cao" }, { "literal": "Yepang Liu" }, { "literal": "Shing-Chi Cheung" }, { "literal": "Hailong Wang" } ], "citation-key": "arxiv_2504_02357", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a646bb2136cdaf47a0d8b3f485d3e1973fc15ac4f844e1beddedf6021b56ffe4", "submitted_at": "2025-04-03T07:45:09Z", "updated_at": "2025-04-03T07:45:09Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.02357v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.02357", "catalog_id": "arxiv:2504.02357", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-03", "identifiers": { "arxiv": "2504.02357" }, "record_updated_at": "2025-04-03" }, "genre": "preprint", "id": "arxiv:2504.02357", "issued": { "date-parts": [ [ 2025, 4, 3 ] ] }, "note": "First published: 2025-04-03; Recorded arXiv version: v1; Recorded version updated: 2025-04-03T07:45:09Z.", "title": "ReuseDroid: A VLM-empowered Android UI Test Migrator Boosted by Active Feedback", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.04716v1", "archive": "arXiv", "archive_location": "2504.04716", "author": [ { "literal": "Haoren Zhao" }, { "literal": "Tianyi Chen" }, { "literal": "Zhen Wang" } ], "citation-key": "arxiv_2504_04716", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ededfa864150f1b7b4ba53b265b5cbaa60bcf42690a8e22f5c13f915b6cd0108", "submitted_at": "2025-04-07T03:58:45Z", "updated_at": "2025-04-07T03:58:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.04716v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.04716", "catalog_id": "arxiv:2504.04716", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-07", "identifiers": { "arxiv": "2504.04716" }, "record_updated_at": "2025-04-07" }, "genre": "preprint", "id": "arxiv:2504.04716", "issued": { "date-parts": [ [ 2025, 4, 7 ] ] }, "note": "First published: 2025-04-07; Recorded arXiv version: v1; Recorded version updated: 2025-04-07T03:58:45Z.", "title": "On the Robustness of GUI Grounding Models Against Image Attacks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.07981v1", "archive": "arXiv", "archive_location": "2504.07981", "author": [ { "literal": "Kaixin Li" }, { "literal": "Ziyang Meng" }, { "literal": "Hongzhan Lin" }, { "literal": "Ziyang Luo" }, { "literal": "Yuchen Tian" }, { "literal": "Jing Ma" }, { "literal": "Zhiyong Huang" }, { "literal": "Tat-Seng Chua" } ], "citation-key": "arxiv_2504_07981", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "679bfcac736544e9a241c579adfda0cacbc40838452ea56ad631f8086868415a", "submitted_at": "2025-04-04T14:25:17Z", "updated_at": "2025-04-04T14:25:17Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.07981v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.07981", "catalog_id": "arxiv:2504.07981", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-04", "identifiers": { "arxiv": "2504.07981" }, "record_updated_at": "2025-04-04" }, "genre": "preprint", "id": "arxiv:2504.07981", "issued": { "date-parts": [ [ 2025, 4, 4 ] ] }, "note": "First published: 2025-04-04; Recorded arXiv version: v1; Recorded version updated: 2025-04-04T14:25:17Z.", "title": "ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.08942v2", "archive": "arXiv", "archive_location": "2504.08942", "author": [ { "literal": "Xing Han Lù" }, { "literal": "Amirhossein Kazemnejad" }, { "literal": "Nicholas Meade" }, { "literal": "Arkil Patel" }, { "literal": "Dongchan Shin" }, { "literal": "Alejandra Zambrano" }, { "literal": "Karolina Stańczak" }, { "literal": "Peter Shaw" }, { "literal": "Christopher J. Pal" }, { "literal": "Siva Reddy" } ], "citation-key": "arxiv_2504_08942", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "53cba694577319ac1992c853be79735d654d9d32c12edcc543e6738ce010050d", "submitted_at": "2025-04-11T19:49:22Z", "updated_at": "2025-10-06T15:46:30Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2504.08942v2" }, "canonical_source_url": "https://arxiv.org/abs/2504.08942", "catalog_id": "arxiv:2504.08942", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-11", "identifiers": { "arxiv": "2504.08942" }, "record_updated_at": "2025-10-06" }, "genre": "preprint", "id": "arxiv:2504.08942", "issued": { "date-parts": [ [ 2025, 4, 11 ] ] }, "note": "First published: 2025-04-11; Recorded arXiv version: v2; Recorded version updated: 2025-10-06T15:46:30Z.", "title": "AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2504.10127v2", "archive": "arXiv", "archive_location": "2504.10127", "author": [ { "literal": "Junlei Zhang" }, { "literal": "Zichen Ding" }, { "literal": "Chang Ma" }, { "literal": "Zijie Chen" }, { "literal": "Qiushi Sun" }, { "literal": "Zhenzhong Lan" }, { "literal": "Junxian He" } ], "citation-key": "arxiv_2504_10127", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "171517999164a3581f71cf363ec6eed357b28ca86908cb97e8498c6958762a5f", "submitted_at": "2025-04-14T11:35:02Z", "updated_at": "2025-04-15T17:13:46Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2504.10127v2" }, "canonical_source_url": "https://arxiv.org/abs/2504.10127", "catalog_id": "arxiv:2504.10127", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-14", "identifiers": { "arxiv": "2504.10127" }, "record_updated_at": "2025-04-15" }, "genre": "preprint", "id": "arxiv:2504.10127", "issued": { "date-parts": [ [ 2025, 4, 14 ] ] }, "note": "First published: 2025-04-14; Recorded arXiv version: v2; Recorded version updated: 2025-04-15T17:13:46Z.", "title": "Breaking the Data Barrier -- Building GUI Agents Through Task Generalization", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2504.10445v2", "archive": "arXiv", "archive_location": "2504.10445", "author": [ { "literal": "Suyu Ye" }, { "literal": "Haojun Shi" }, { "literal": "Darren Shih" }, { "literal": "Hyokun Yun" }, { "literal": "Tanya Roosta" }, { "literal": "Tianmin Shu" } ], "citation-key": "arxiv_2504_10445", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "268adf347e1f224f434e6dba838873c99a4050686cae1495643d9638d840d809", "submitted_at": "2025-04-14T17:36:46Z", "updated_at": "2025-12-01T15:22:40Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2504.10445v2" }, "canonical_source_url": "https://arxiv.org/abs/2504.10445", "catalog_id": "arxiv:2504.10445", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-14", "identifiers": { "arxiv": "2504.10445" }, "record_updated_at": "2025-12-01" }, "genre": "preprint", "id": "arxiv:2504.10445", "issued": { "date-parts": [ [ 2025, 4, 14 ] ] }, "note": "First published: 2025-04-14; Recorded arXiv version: v2; Recorded version updated: 2025-12-01T15:22:40Z.", "title": "RealWebAssist: A Benchmark for Long-Horizon Web Assistance with Real-World Users", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2504.11257v4", "archive": "arXiv", "archive_location": "2504.11257", "author": [ { "literal": "Xinyi Liu" }, { "literal": "Xiaoyi Zhang" }, { "literal": "Ziyun Zhang" }, { "literal": "Yan Lu" } ], "citation-key": "arxiv_2504_11257", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "363e3c3e882fa93a19df74bbf77aa98525cf4639c8f62100c3ddc3e0a438997f", "submitted_at": "2025-04-15T14:56:21Z", "updated_at": "2025-07-30T13:57:49Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2504.11257v4" }, "canonical_source_url": "https://arxiv.org/abs/2504.11257", "catalog_id": "arxiv:2504.11257", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "record", "first_published_at": "2025-04-15", "identifiers": { "arxiv": "2504.11257" }, "publication": { "authors": [ "Xinyi Liu", "Xiaoyi Zhang", "Ziyun Zhang", "Yan Lu" ], "date_basis": "Explicit publisher citation_publication_date: 2025/7. Preserve supplied precision; do not substitute conference event dates.", "date_parts": [ 2025, 7 ], "doi": "10.18653/v1/2025.findings-acl.809", "event_dates": null, "evidence_urls": [ "https://aclanthology.org/2025.findings-acl.809/" ], "identity_basis": "Same work: normalized title, ordered author correspondence and the specific contribution in the complete publisher abstract were reviewed against the retained arXiv identity.", "issue": null, "notes": "Publisher proceedings edition; retained separately from the arXiv snapshot.", "pages": "15668-15684", "pdf_url": "https://aclanthology.org/2025.findings-acl.809.pdf", "publisher": "Association for Computational Linguistics", "series": null, "title": "UI-E2I-Synth: Advancing GUI Grounding with Large-Scale Instruction Synthesis", "type": "conference-paper", "url": "https://aclanthology.org/2025.findings-acl.809/", "venue": "Findings of the Association for Computational Linguistics: ACL 2025", "volume": null }, "record_updated_at": "2025-07-30" }, "genre": "preprint", "id": "arxiv:2504.11257", "issued": { "date-parts": [ [ 2025, 4, 15 ] ] }, "note": "First published: 2025-04-15; Recorded arXiv version: v4; Recorded version updated: 2025-07-30T13:57:49Z.", "title": "UI-E2I-Synth: Advancing GUI Grounding with Large-Scale Instruction Synthesis", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2504.11281v1", "archive": "arXiv", "archive_location": "2504.11281", "author": [ { "literal": "Chaoran Chen" }, { "literal": "Zhiping Zhang" }, { "literal": "Bingcan Guo" }, { "literal": "Shang Ma" }, { "literal": "Ibrahim Khalilov" }, { "literal": "Simret A Gebreegziabher" }, { "literal": "Yanfang Ye" }, { "literal": "Ziang Xiao" }, { "literal": "Yaxing Yao" }, { "literal": "Tianshi Li" }, { "literal": "Toby Jia-Jun Li" } ], "citation-key": "arxiv_2504_11281", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "48acf3279b854d45ec076e630e5a84601879fc9ff894c5d368adb59d03bd3de6", "submitted_at": "2025-04-15T15:21:09Z", "updated_at": "2025-04-15T15:21:09Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.11281v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.11281", "catalog_id": "arxiv:2504.11281", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-15", "identifiers": { "arxiv": "2504.11281" }, "record_updated_at": "2025-04-15" }, "genre": "preprint", "id": "arxiv:2504.11281", "issued": { "date-parts": [ [ 2025, 4, 15 ] ] }, "note": "First published: 2025-04-15; Recorded arXiv version: v1; Recorded version updated: 2025-04-15T15:21:09Z.", "title": "The Obvious Invisible Threat: LLM-Powered GUI Agents' Vulnerability to Fine-Print Injections", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.11675v1", "archive": "arXiv", "archive_location": "2504.11675", "author": [ { "literal": "Biniam Fisseha Demissie" }, { "literal": "Yan Naing Tun" }, { "literal": "Lwin Khin Shar" }, { "literal": "Mariano Ceccato" } ], "citation-key": "arxiv_2504_11675", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1a25cb3ca680732a08cf2c461a65caa0c6a70404224bdfd919ba5bec7760ba72", "submitted_at": "2025-04-16T00:19:31Z", "updated_at": "2025-04-16T00:19:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.11675v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.11675", "catalog_id": "arxiv:2504.11675", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-16", "identifiers": { "arxiv": "2504.11675" }, "record_updated_at": "2025-04-16" }, "genre": "preprint", "id": "arxiv:2504.11675", "issued": { "date-parts": [ [ 2025, 4, 16 ] ] }, "note": "First published: 2025-04-16; Recorded arXiv version: v1; Recorded version updated: 2025-04-16T00:19:31Z.", "title": "VLM-Fuzz: Vision Language Model Assisted Recursive Depth-first Search Exploration for Effective UI Testing of Android Apps", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.13805v1", "archive": "arXiv", "archive_location": "2504.13805", "author": [ { "literal": "Guangyi Liu" }, { "literal": "Pengxiang Zhao" }, { "literal": "Liang Liu" }, { "literal": "Zhiming Chen" }, { "literal": "Yuxiang Chai" }, { "literal": "Shuai Ren" }, { "literal": "Hao Wang" }, { "literal": "Shibo He" }, { "literal": "Wenchao Meng" } ], "citation-key": "arxiv_2504_13805", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2bbe6c4969a9557546752235809d25912bff96a97001fade30102d8d2c751054", "submitted_at": "2025-04-18T17:13:34Z", "updated_at": "2025-04-18T17:13:34Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2504.13805v1" }, "canonical_source_url": "https://arxiv.org/abs/2504.13805", "catalog_id": "arxiv:2504.13805", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-18", "identifiers": { "arxiv": "2504.13805" }, "record_updated_at": "2025-04-18" }, "genre": "preprint", "id": "arxiv:2504.13805", "issued": { "date-parts": [ [ 2025, 4, 18 ] ] }, "note": "First published: 2025-04-18; Recorded arXiv version: v1; Recorded version updated: 2025-04-18T17:13:34Z.", "title": "LearnAct: Few-Shot Mobile GUI Agent with a Unified Demonstration Benchmark", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2504.14064v3", "archive": "arXiv", "archive_location": "2504.14064", "author": [ { "literal": "Leo Boisvert" }, { "literal": "Mihir Bansal" }, { "literal": "Chandra Kiran Reddy Evuru" }, { "literal": "Gabriel Huang" }, { "literal": "Abhay Puri" }, { "literal": "Avinandan Bose" }, { "literal": "Maryam Fazel" }, { "literal": "Quentin Cappart" }, { "literal": "Jason Stanley" }, { "literal": "Alexandre Lacoste" }, { "literal": "Alexandre Drouin" }, { "literal": "Krishnamurthy Dvijotham" } ], "citation-key": "arxiv_2504_14064", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c5bdf5fefa1ee31126d99de08236991df559896212f60e0b3058fcd8cc8a134b", "submitted_at": "2025-04-18T20:36:10Z", "updated_at": "2025-10-07T13:10:30Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2504.14064v3" }, "canonical_source_url": "https://arxiv.org/abs/2504.14064", "catalog_id": "arxiv:2504.14064", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-18", "identifiers": { "arxiv": "2504.14064" }, "record_updated_at": "2025-10-07" }, "genre": "preprint", "id": "arxiv:2504.14064", "issued": { "date-parts": [ [ 2025, 4, 18 ] ] }, "note": "First published: 2025-04-18; Recorded arXiv version: v3; Recorded version updated: 2025-10-07T13:10:30Z.", "title": "DoomArena: A framework for Testing AI Agents Against Evolving Security Threats", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2504.17934v2", "archive": "arXiv", "archive_location": "2504.17934", "author": [ { "literal": "Chaoran Chen" }, { "literal": "Zhiping Zhang" }, { "literal": "Ibrahim Khalilov" }, { "literal": "Bingcan Guo" }, { "literal": "Simret A Gebreegziabher" }, { "literal": "Yanfang Ye" }, { "literal": "Ziang Xiao" }, { "literal": "Yaxing Yao" }, { "literal": "Tianshi Li" }, { "literal": "Toby Jia-Jun Li" } ], "citation-key": "arxiv_2504_17934", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d2a55e347bc10e5571464924d9301ff2c32d5e9f1f990ca91725ec8c9f6da594", "submitted_at": "2025-04-24T20:51:20Z", "updated_at": "2025-06-05T12:40:53Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2504.17934v2" }, "canonical_source_url": "https://arxiv.org/abs/2504.17934", "catalog_id": "arxiv:2504.17934", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-24", "identifiers": { "arxiv": "2504.17934" }, "record_updated_at": "2025-06-05" }, "genre": "preprint", "id": "arxiv:2504.17934", "issued": { "date-parts": [ [ 2025, 4, 24 ] ] }, "note": "First published: 2025-04-24; Recorded arXiv version: v2; Recorded version updated: 2025-06-05T12:40:53Z.", "title": "Toward a Human-Centered Evaluation Framework for Trustworthy LLM-Powered GUI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2504.18575v3", "archive": "arXiv", "archive_location": "2504.18575", "author": [ { "literal": "Ivan Evtimov" }, { "literal": "Arman Zharmagambetov" }, { "literal": "Aaron Grattafiori" }, { "literal": "Chuan Guo" }, { "literal": "Kamalika Chaudhuri" } ], "citation-key": "arxiv_2504_18575", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3a104d9076797555795766749f6b4eeec9428d0efb9340f7e55cffecd5d727b0", "submitted_at": "2025-04-22T17:51:03Z", "updated_at": "2025-05-16T22:42:29Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2504.18575v3" }, "canonical_source_url": "https://arxiv.org/abs/2504.18575", "catalog_id": "arxiv:2504.18575", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-04-22", "identifiers": { "arxiv": "2504.18575" }, "record_updated_at": "2025-05-16" }, "genre": "preprint", "id": "arxiv:2504.18575", "issued": { "date-parts": [ [ 2025, 4, 22 ] ] }, "note": "First published: 2025-04-22; Recorded arXiv version: v3; Recorded version updated: 2025-05-16T22:42:29Z.", "title": "WASP: Benchmarking Web Agent Security Against Prompt Injection Attacks", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2505.03570v1", "archive": "arXiv", "archive_location": "2505.03570", "author": [ { "literal": "Mariya Davydova" }, { "literal": "Daniel Jeffries" }, { "literal": "Patrick Barker" }, { "literal": "Arturo Márquez Flores" }, { "literal": "Sinéad Ryan" } ], "citation-key": "arxiv_2505_03570", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e8bb4c6f8080ed1ee91e1089233cae6f231ab3256b38ac7d6403ebea3ae56fe0", "submitted_at": "2025-05-06T14:29:47Z", "updated_at": "2025-05-06T14:29:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2505.03570v1" }, "canonical_source_url": "https://arxiv.org/abs/2505.03570", "catalog_id": "arxiv:2505.03570", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-06", "identifiers": { "arxiv": "2505.03570" }, "record_updated_at": "2025-05-06" }, "genre": "preprint", "id": "arxiv:2505.03570", "issued": { "date-parts": [ [ 2025, 5, 6 ] ] }, "note": "First published: 2025-05-06; Recorded arXiv version: v1; Recorded version updated: 2025-05-06T14:29:47Z.", "title": "OSUniverse: Benchmark for Multimodal GUI-navigation AI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2505.05026v5", "archive": "arXiv", "archive_location": "2505.05026", "author": [ { "literal": "Jaehyun Jeon" }, { "literal": "Min Soo Kim" }, { "literal": "Jang Han Yoon" }, { "literal": "Sumin Shim" }, { "literal": "Yejin Choi" }, { "literal": "Hanbin Kim" }, { "literal": "Dae Hyun Kim" }, { "literal": "Youngjae Yu" } ], "citation-key": "arxiv_2505_05026", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3f65cd2856c554584baab9b1cbf9c5250b5c01d4f31980d69aefab727427ded2", "submitted_at": "2025-05-08T08:00:32Z", "updated_at": "2026-06-04T06:58:13Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2505.05026v5" }, "canonical_source_url": "https://arxiv.org/abs/2505.05026", "catalog_id": "arxiv:2505.05026", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2025-05-08", "identifiers": { "arxiv": "2505.05026" }, "record_updated_at": "2026-06-04" }, "genre": "preprint", "id": "arxiv:2505.05026", "issued": { "date-parts": [ [ 2025, 5, 8 ] ] }, "note": "First published: 2025-05-08; Recorded arXiv version: v5; Recorded version updated: 2026-06-04T06:58:13Z.", "title": "Do MLLMs Capture How Interfaces Guide User Behavior? A Benchmark for Multimodal UI/UX Design Understanding", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2505.09875v2", "archive": "arXiv", "archive_location": "2505.09875", "author": [ { "literal": "Shuning Zhang" }, { "literal": "Jingruo Chen" }, { "literal": "Zhiqi Gao" }, { "literal": "Jiajing Gao" }, { "literal": "Xin Yi" }, { "literal": "Hewu Li" } ], "citation-key": "arxiv_2505_09875", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "63f4a505645fec16eac6b59923a6c1979c53376f19c6a4ea53c9a5d32a45e3d4", "submitted_at": "2025-05-15T00:42:51Z", "updated_at": "2025-05-16T07:01:20Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2505.09875v2" }, "canonical_source_url": "https://arxiv.org/abs/2505.09875", "catalog_id": "arxiv:2505.09875", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-15", "identifiers": { "arxiv": "2505.09875" }, "record_updated_at": "2025-05-16" }, "genre": "preprint", "id": "arxiv:2505.09875", "issued": { "date-parts": [ [ 2025, 5, 15 ] ] }, "note": "First published: 2025-05-15; Recorded arXiv version: v2; Recorded version updated: 2025-05-16T07:01:20Z.", "title": "Characterizing Unintended Consequences in Human-GUI Agent Collaboration for Web Browsing", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2505.11891v3", "archive": "arXiv", "archive_location": "2505.11891", "author": [ { "literal": "Weikai Xu" }, { "literal": "Zhizheng Jiang" }, { "literal": "Yuxuan Liu" }, { "literal": "Pengzhi Gao" }, { "literal": "Wei Liu" }, { "literal": "Jian Luan" }, { "literal": "Yuanchun Li" }, { "literal": "Yunxin Liu" }, { "literal": "Bin Wang" }, { "literal": "Bo An" } ], "citation-key": "arxiv_2505_11891", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5ca1ebfdd5a96e146d37539294462c7fb67bf20deaf77307de37aa3bb1dde082", "submitted_at": "2025-05-17T07:58:34Z", "updated_at": "2026-02-02T06:55:10Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2505.11891v3" }, "canonical_source_url": "https://arxiv.org/abs/2505.11891", "catalog_id": "arxiv:2505.11891", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-17", "identifiers": { "arxiv": "2505.11891" }, "record_updated_at": "2026-02-02" }, "genre": "preprint", "id": "arxiv:2505.11891", "issued": { "date-parts": [ [ 2025, 5, 17 ] ] }, "note": "First published: 2025-05-17; Recorded arXiv version: v3; Recorded version updated: 2026-02-02T06:55:10Z.", "title": "Mobile-Bench-v2: A More Realistic and Comprehensive Benchmark for VLM-based Mobile Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2505.12842v4", "archive": "arXiv", "archive_location": "2505.12842", "author": [ { "literal": "Zheng Wu" }, { "literal": "Pengzhou Cheng" }, { "literal": "Zongru Wu" }, { "literal": "Lingzhong Dong" }, { "literal": "Zhuosheng Zhang" } ], "citation-key": "arxiv_2505_12842", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ce9570eb6a52feb220cdd2eb4c7f37a2ea63082febe48888a2654c974138a82d", "submitted_at": "2025-05-19T08:29:05Z", "updated_at": "2025-11-22T15:07:27Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2505.12842v4" }, "canonical_source_url": "https://arxiv.org/abs/2505.12842", "catalog_id": "arxiv:2505.12842", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-19", "identifiers": { "arxiv": "2505.12842" }, "record_updated_at": "2025-11-22" }, "genre": "preprint", "id": "arxiv:2505.12842", "issued": { "date-parts": [ [ 2025, 5, 19 ] ] }, "note": "First published: 2025-05-19; Recorded arXiv version: v4; Recorded version updated: 2025-11-22T15:07:27Z.", "title": "GEM: Gaussian Embedding Modeling for Out-of-Distribution Detection in GUI Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2505.12981v2", "archive": "arXiv", "archive_location": "2505.12981", "author": [ { "literal": "Liangxuan Wu" }, { "literal": "Chao Wang" }, { "literal": "Tianming Liu" }, { "literal": "Yanjie Zhao" }, { "literal": "Haoyu Wang" } ], "citation-key": "arxiv_2505_12981", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9657a66aebe46151ed56bf76131ffe45f52459c1cf0e000907113776c91259d9", "submitted_at": "2025-05-19T11:17:46Z", "updated_at": "2025-05-20T07:02:05Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2505.12981v2" }, "canonical_source_url": "https://arxiv.org/abs/2505.12981", "catalog_id": "arxiv:2505.12981", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-19", "identifiers": { "arxiv": "2505.12981" }, "record_updated_at": "2025-05-20" }, "genre": "preprint", "id": "arxiv:2505.12981", "issued": { "date-parts": [ [ 2025, 5, 19 ] ] }, "note": "First published: 2025-05-19; Recorded arXiv version: v2; Recorded version updated: 2025-05-20T07:02:05Z.", "title": "From Assistants to Adversaries: Exploring the Security Risks of Mobile LLM Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2505.13076v1", "archive": "arXiv", "archive_location": "2505.13076", "author": [ { "literal": "Mykyta Mudryi" }, { "literal": "Markiyan Chaklosh" }, { "literal": "Grzegorz Wójcik" } ], "citation-key": "arxiv_2505_13076", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "709ee511ffa39e805fcc58abe066cf2cec6c1707a4a9c64d9cfd107bef7544af", "submitted_at": "2025-05-19T13:10:29Z", "updated_at": "2025-05-19T13:10:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2505.13076v1" }, "canonical_source_url": "https://arxiv.org/abs/2505.13076", "catalog_id": "arxiv:2505.13076", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-19", "identifiers": { "arxiv": "2505.13076" }, "record_updated_at": "2025-05-19" }, "genre": "preprint", "id": "arxiv:2505.13076", "issued": { "date-parts": [ [ 2025, 5, 19 ] ] }, "note": "First published: 2025-05-19; Recorded arXiv version: v1; Recorded version updated: 2025-05-19T13:10:29Z.", "title": "The Hidden Dangers of Browsing AI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2505.13227v3", "archive": "arXiv", "archive_location": "2505.13227", "author": [ { "literal": "Tianbao Xie" }, { "literal": "Jiaqi Deng" }, { "literal": "Xiaochuan Li" }, { "literal": "Junlin Yang" }, { "literal": "Haoyuan Wu" }, { "literal": "Jixuan Chen" }, { "literal": "Wenjing Hu" }, { "literal": "Xinyuan Wang" }, { "literal": "Yuhui Xu" }, { "literal": "Zekun Wang" }, { "literal": "Yiheng Xu" }, { "literal": "Junli Wang" }, { "literal": "Doyen Sahoo" }, { "literal": "Tao Yu" }, { "literal": "Caiming Xiong" } ], "citation-key": "arxiv_2505_13227", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1bec4ad578c01ebdfb58bd269562d3ca7a2bd8d2687cde07cb867db0b78cb105", "submitted_at": "2025-05-19T15:09:23Z", "updated_at": "2025-10-24T19:08:03Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2505.13227v3" }, "canonical_source_url": "https://arxiv.org/abs/2505.13227", "catalog_id": "arxiv:2505.13227", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-19", "identifiers": { "arxiv": "2505.13227" }, "record_updated_at": "2025-10-24" }, "genre": "preprint", "id": "arxiv:2505.13227", "issued": { "date-parts": [ [ 2025, 5, 19 ] ] }, "note": "First published: 2025-05-19; Recorded arXiv version: v3; Recorded version updated: 2025-10-24T19:08:03Z.", "title": "Scaling Computer-Use Grounding via User Interface Decomposition and Synthesis", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2505.13909v2", "archive": "arXiv", "archive_location": "2505.13909", "author": [ { "literal": "Yanheng He" }, { "literal": "Jiahe Jin" }, { "literal": "Pengfei Liu" } ], "citation-key": "arxiv_2505_13909", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "026a64046926912a54e6a7c6cf62a17740a6f89a94713615f9750f8bf8302c78", "submitted_at": "2025-05-20T04:20:18Z", "updated_at": "2026-03-03T03:55:55Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2505.13909v2" }, "canonical_source_url": "https://arxiv.org/abs/2505.13909", "catalog_id": "arxiv:2505.13909", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-20", "identifiers": { "arxiv": "2505.13909" }, "record_updated_at": "2026-03-03" }, "genre": "preprint", "id": "arxiv:2505.13909", "issued": { "date-parts": [ [ 2025, 5, 20 ] ] }, "note": "First published: 2025-05-20; Recorded arXiv version: v2; Recorded version updated: 2026-03-03T03:55:55Z.", "title": "Efficient Agent Training for Computer Use", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2505.14289v2", "archive": "arXiv", "archive_location": "2505.14289", "author": [ { "literal": "Yijie Lu" }, { "literal": "Manman Zhao" }, { "literal": "Tianjie Ju" }, { "literal": "Zihe Yan" }, { "literal": "Xinbei Ma" }, { "literal": "Yuan Guo" }, { "literal": "Daizong Ding" }, { "literal": "Gongshen Liu" }, { "literal": "Zhuosheng Zhang" } ], "citation-key": "arxiv_2505_14289", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2c256d8b0d774efdeee896860f8c504c0d1593215508fdf0881d3fd3cdc3968c", "submitted_at": "2025-05-20T12:41:05Z", "updated_at": "2026-06-05T10:24:33Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2505.14289v2" }, "canonical_source_url": "https://arxiv.org/abs/2505.14289", "catalog_id": "arxiv:2505.14289", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-20", "identifiers": { "arxiv": "2505.14289" }, "record_updated_at": "2026-06-05" }, "genre": "preprint", "id": "arxiv:2505.14289", "issued": { "date-parts": [ [ 2025, 5, 20 ] ] }, "note": "First published: 2025-05-20; Recorded arXiv version: v2; Recorded version updated: 2026-06-05T10:24:33Z.", "title": "EVA: Evolving Semantic Adversaries for Red-Teaming GUI Agents Against Environmental Injection Attacks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2505.15277v2", "archive": "arXiv", "archive_location": "2505.15277", "author": [ { "literal": "Hyungjoo Chae" }, { "literal": "Sunghwan Kim" }, { "literal": "Junhee Cho" }, { "literal": "Seungone Kim" }, { "literal": "Seungjun Moon" }, { "literal": "Gyeom Hwangbo" }, { "literal": "Dongha Lim" }, { "literal": "Minjin Kim" }, { "literal": "Yeonjun Hwang" }, { "literal": "Minju Gwak" }, { "literal": "Dongwook Choi" }, { "literal": "Minseok Kang" }, { "literal": "Gwanhoon Im" }, { "literal": "ByeongUng Cho" }, { "literal": "Hyojun Kim" }, { "literal": "Jun Hee Han" }, { "literal": "Taeyoon Kwon" }, { "literal": "Minju Kim" }, { "literal": "Beong-woo Kwak" }, { "literal": "Dongjin Kang" }, { "literal": "Jinyoung Yeo" } ], "citation-key": "arxiv_2505_15277", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "440abdc75c0a51478cfedc6ee0a9196ccff641ec5a23849ba56be43448bab15f", "submitted_at": "2025-05-21T08:56:55Z", "updated_at": "2025-11-25T21:34:57Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2505.15277v2" }, "canonical_source_url": "https://arxiv.org/abs/2505.15277", "catalog_id": "arxiv:2505.15277", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-21", "identifiers": { "arxiv": "2505.15277" }, "record_updated_at": "2025-11-25" }, "genre": "preprint", "id": "arxiv:2505.15277", "issued": { "date-parts": [ [ 2025, 5, 21 ] ] }, "note": "First published: 2025-05-21; Recorded arXiv version: v2; Recorded version updated: 2025-11-25T21:34:57Z.", "title": "Web-Shepherd: Advancing PRMs for Reinforcing Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2505.19897v3", "archive": "arXiv", "archive_location": "2505.19897", "author": [ { "literal": "Qiushi Sun" }, { "literal": "Zhoumianze Liu" }, { "literal": "Chang Ma" }, { "literal": "Zichen Ding" }, { "literal": "Fangzhi Xu" }, { "literal": "Zhangyue Yin" }, { "literal": "Haiteng Zhao" }, { "literal": "Zhenyu Wu" }, { "literal": "Kanzhi Cheng" }, { "literal": "Zhaoyang Liu" }, { "literal": "Jianing Wang" }, { "literal": "Qintong Li" }, { "literal": "Xiangru Tang" }, { "literal": "Tianbao Xie" }, { "literal": "Xiachong Feng" }, { "literal": "Xiang Li" }, { "literal": "Ben Kao" }, { "literal": "Wenhai Wang" }, { "literal": "Biqing Qi" }, { "literal": "Lingpeng Kong" }, { "literal": "Zhiyong Wu" } ], "citation-key": "arxiv_2505_19897", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4743f6bfd0ca48717184cf55545e33a229dd5ab2d454385d33ea2f1b25c15b0c", "submitted_at": "2025-05-26T12:27:27Z", "updated_at": "2026-04-20T16:29:04Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2505.19897v3" }, "canonical_source_url": "https://arxiv.org/abs/2505.19897", "catalog_id": "arxiv:2505.19897", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-26", "identifiers": { "arxiv": "2505.19897" }, "record_updated_at": "2026-04-20" }, "genre": "preprint", "id": "arxiv:2505.19897", "issued": { "date-parts": [ [ 2025, 5, 26 ] ] }, "note": "First published: 2025-05-26; Recorded arXiv version: v3; Recorded version updated: 2026-04-20T16:29:04Z.", "title": "ScienceBoard: Evaluating Multimodal Autonomous Agents in Realistic Scientific Workflows", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2505.21279v2", "archive": "arXiv", "archive_location": "2505.21279", "author": [ { "literal": "Shaoqing Zhang" }, { "literal": "Kehai Chen" }, { "literal": "Zhuosheng Zhang" }, { "literal": "Rumei Li" }, { "literal": "Rongxiang Weng" }, { "literal": "Yang Xiang" }, { "literal": "Min Zhang" } ], "citation-key": "arxiv_2505_21279", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4fe40099905712f9d7ba89bc7616d09b260fe6113beb8122cade54a66d46e3d7", "submitted_at": "2025-05-27T14:49:30Z", "updated_at": "2025-09-26T09:13:44Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2505.21279v2" }, "canonical_source_url": "https://arxiv.org/abs/2505.21279", "catalog_id": "arxiv:2505.21279", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-27", "identifiers": { "arxiv": "2505.21279" }, "record_updated_at": "2025-09-26" }, "genre": "preprint", "id": "arxiv:2505.21279", "issued": { "date-parts": [ [ 2025, 5, 27 ] ] }, "note": "First published: 2025-05-27; Recorded arXiv version: v2; Recorded version updated: 2025-09-26T09:13:44Z.", "title": "XBOUND: Exploring Capability Boundaries of Device-Control Agents at the State Level", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2505.21499v1", "archive": "arXiv", "archive_location": "2505.21499", "author": [ { "literal": "Haowei Wang" }, { "literal": "Junjie Wang" }, { "literal": "Xiaojun Jia" }, { "literal": "Rupeng Zhang" }, { "literal": "Mingyang Li" }, { "literal": "Zhe Liu" }, { "literal": "Yang Liu" }, { "literal": "Qing Wang" } ], "citation-key": "arxiv_2505_21499", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f8aade17e67471c87f259d49aef43b3caf764b6ac14f99e1a4114f4a25dda3d6", "submitted_at": "2025-05-27T17:59:05Z", "updated_at": "2025-05-27T17:59:05Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2505.21499v1" }, "canonical_source_url": "https://arxiv.org/abs/2505.21499", "catalog_id": "arxiv:2505.21499", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-27", "identifiers": { "arxiv": "2505.21499" }, "record_updated_at": "2025-05-27" }, "genre": "preprint", "id": "arxiv:2505.21499", "issued": { "date-parts": [ [ 2025, 5, 27 ] ] }, "note": "First published: 2025-05-27; Recorded arXiv version: v1; Recorded version updated: 2025-05-27T17:59:05Z.", "title": "AdInject: Real-World Black-Box Attacks on Web Agents via Advertising Delivery", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2505.21936v5", "archive": "arXiv", "archive_location": "2505.21936", "author": [ { "literal": "Zeyi Liao" }, { "literal": "Jaylen Jones" }, { "literal": "Linxi Jiang" }, { "literal": "Yuting Ning" }, { "literal": "Eric Fosler-Lussier" }, { "literal": "Yu Su" }, { "literal": "Zhiqiang Lin" }, { "literal": "Huan Sun" } ], "citation-key": "arxiv_2505_21936", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7fe3bdb1357a375de0b1700a6eda7d5987f358bd47fddf3405c27272a3453f2f", "submitted_at": "2025-05-28T03:42:09Z", "updated_at": "2026-03-01T07:50:09Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2505.21936v5" }, "canonical_source_url": "https://arxiv.org/abs/2505.21936", "catalog_id": "arxiv:2505.21936", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-28", "identifiers": { "arxiv": "2505.21936" }, "record_updated_at": "2026-03-01" }, "genre": "preprint", "id": "arxiv:2505.21936", "issued": { "date-parts": [ [ 2025, 5, 28 ] ] }, "note": "First published: 2025-05-28; Recorded arXiv version: v5; Recorded version updated: 2026-03-01T07:50:09Z.", "title": "RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environments", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2505.24878v1", "archive": "arXiv", "archive_location": "2505.24878", "author": [ { "literal": "Yaxin Luo" }, { "literal": "Zhaoyi Li" }, { "literal": "Jiacheng Liu" }, { "literal": "Jiacheng Cui" }, { "literal": "Xiaohan Zhao" }, { "literal": "Zhiqiang Shen" } ], "citation-key": "arxiv_2505_24878", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3f2206333ca25b3f21df7ce49adfd8bc56521b9b8ace7a9b686eccf5b67276fa", "submitted_at": "2025-05-30T17:59:55Z", "updated_at": "2025-05-30T17:59:55Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2505.24878v1" }, "canonical_source_url": "https://arxiv.org/abs/2505.24878", "catalog_id": "arxiv:2505.24878", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-30", "identifiers": { "arxiv": "2505.24878" }, "record_updated_at": "2025-05-30" }, "genre": "preprint", "id": "arxiv:2505.24878", "issued": { "date-parts": [ [ 2025, 5, 30 ] ] }, "note": "First published: 2025-05-30; Recorded arXiv version: v1; Recorded version updated: 2025-05-30T17:59:55Z.", "title": "Open CaptchaWorld: A Comprehensive Web-based Platform for Testing and Benchmarking Multimodal LLM Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.00618v3", "archive": "arXiv", "archive_location": "2506.00618", "author": [ { "literal": "Jingyi Yang" }, { "literal": "Shuai Shao" }, { "literal": "Dongrui Liu" }, { "literal": "Jing Shao" } ], "citation-key": "arxiv_2506_00618", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bb94fdba775cb0fb54b312fdff9fa71afd3aefa56b812ff31ceccad33eb5f800", "submitted_at": "2025-05-31T16:04:59Z", "updated_at": "2025-06-20T01:24:06Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2506.00618v3" }, "canonical_source_url": "https://arxiv.org/abs/2506.00618", "catalog_id": "arxiv:2506.00618", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-05-31", "identifiers": { "arxiv": "2506.00618" }, "record_updated_at": "2025-06-20" }, "genre": "preprint", "id": "arxiv:2506.00618", "issued": { "date-parts": [ [ 2025, 5, 31 ] ] }, "note": "First published: 2025-05-31; Recorded arXiv version: v3; Recorded version updated: 2025-06-20T01:24:06Z.", "title": "RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2506.01616v1", "archive": "arXiv", "archive_location": "2506.01616", "author": [ { "literal": "Xiao Yang" }, { "literal": "Jiawei Chen" }, { "literal": "Jun Luo" }, { "literal": "Zhengwei Fang" }, { "literal": "Yinpeng Dong" }, { "literal": "Hang Su" }, { "literal": "Jun Zhu" } ], "citation-key": "arxiv_2506_01616", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1ec2467703bc5d0f97ef8d207289ed56f6d12b811688d20b7833eac06038757b", "submitted_at": "2025-06-02T12:56:27Z", "updated_at": "2025-06-02T12:56:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.01616v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.01616", "catalog_id": "arxiv:2506.01616", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-02", "identifiers": { "arxiv": "2506.01616" }, "record_updated_at": "2025-06-02" }, "genre": "preprint", "id": "arxiv:2506.01616", "issued": { "date-parts": [ [ 2025, 6, 2 ] ] }, "note": "First published: 2025-06-02; Recorded arXiv version: v1; Recorded version updated: 2025-06-02T12:56:27Z.", "title": "MLA-Trust: Benchmarking Trustworthiness of Multimodal LLM Agents in GUI Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.02456v2", "archive": "arXiv", "archive_location": "2506.02456", "author": [ { "literal": "Tri Cao" }, { "literal": "Bennett Lim" }, { "literal": "Yue Liu" }, { "literal": "Yuan Sui" }, { "literal": "Yuexin Li" }, { "literal": "Shumin Deng" }, { "literal": "Lin Lu" }, { "literal": "Nay Oo" }, { "literal": "Shuicheng Yan" }, { "literal": "Bryan Hooi" } ], "citation-key": "arxiv_2506_02456", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3cd988217d1e3371bec827b472abf46c205ac407b5b60b7b581d7105f86e8bc5", "submitted_at": "2025-06-03T05:21:50Z", "updated_at": "2026-03-01T16:12:40Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.02456v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.02456", "catalog_id": "arxiv:2506.02456", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-03", "identifiers": { "arxiv": "2506.02456" }, "record_updated_at": "2026-03-01" }, "genre": "preprint", "id": "arxiv:2506.02456", "issued": { "date-parts": [ [ 2025, 6, 3 ] ] }, "note": "First published: 2025-06-03; Recorded arXiv version: v2; Recorded version updated: 2026-03-01T16:12:40Z.", "title": "VPI-Bench: Visual Prompt Injection Attacks for Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2506.02839v1", "archive": "arXiv", "archive_location": "2506.02839", "author": [ { "literal": "Yougang Lyu" }, { "literal": "Xiaoyu Zhang" }, { "literal": "Lingyong Yan" }, { "literal": "Maarten de Rijke" }, { "literal": "Zhaochun Ren" }, { "literal": "Xiuying Chen" } ], "citation-key": "arxiv_2506_02839", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bebdf1c8f34582408ef0050245f6b1e454d1bd2e75903a4e0817a4f3c00fb4db", "submitted_at": "2025-06-03T13:08:17Z", "updated_at": "2025-06-03T13:08:17Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.02839v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.02839", "catalog_id": "arxiv:2506.02839", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-03", "identifiers": { "arxiv": "2506.02839" }, "record_updated_at": "2025-06-03" }, "genre": "preprint", "id": "arxiv:2506.02839", "issued": { "date-parts": [ [ 2025, 6, 3 ] ] }, "note": "First published: 2025-06-03; Recorded arXiv version: v1; Recorded version updated: 2025-06-03T13:08:17Z.", "title": "DeepShop: A Benchmark for Deep Research Shopping Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.02865v2", "archive": "arXiv", "archive_location": "2506.02865", "author": [ { "literal": "Mathieu Andreux" }, { "literal": "Breno Baldas Skuk" }, { "literal": "Hamza Benchekroun" }, { "literal": "Emilien Biré" }, { "literal": "Antoine Bonnet" }, { "literal": "Riaz Bordie" }, { "literal": "Nathan Bout" }, { "literal": "Matthias Brunel" }, { "literal": "Pierre-Louis Cedoz" }, { "literal": "Antoine Chassang" }, { "literal": "Mickaël Chen" }, { "literal": "Alexandra D. Constantinou" }, { "literal": "Antoine d'Andigné" }, { "literal": "Hubert de La Jonquière" }, { "literal": "Aurélien Delfosse" }, { "literal": "Ludovic Denoyer" }, { "literal": "Alexis Deprez" }, { "literal": "Augustin Derupti" }, { "literal": "Michael Eickenberg" }, { "literal": "Mathïs Federico" }, { "literal": "Charles Kantor" }, { "literal": "Xavier Koegler" }, { "literal": "Yann Labbé" }, { "literal": "Matthew C. H. Lee" }, { "literal": "Erwan Le Jumeau de Kergaradec" }, { "literal": "Amir Mahla" }, { "literal": "Avshalom Manevich" }, { "literal": "Adrien Maret" }, { "literal": "Charles Masson" }, { "literal": "Rafaël Maurin" }, { "literal": "Arturo Mena" }, { "literal": "Philippe Modard" }, { "literal": "Axel Moyal" }, { "literal": "Axel Nguyen Kerbel" }, { "literal": "Julien Revelle" }, { "literal": "Mats L. Richter" }, { "literal": "María Santos" }, { "literal": "Laurent Sifre" }, { "literal": "Maxime Theillard" }, { "literal": "Marc Thibault" }, { "literal": "Louis Thiry" }, { "literal": "Léo Tronchon" }, { "literal": "Nicolas Usunier" }, { "literal": "Tony Wu" } ], "citation-key": "arxiv_2506_02865", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "89c2d658e03204b56f6d273da9572820f9a23ca359a441ea87baa451a93c5b58", "submitted_at": "2025-06-03T13:29:03Z", "updated_at": "2025-06-11T09:33:38Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.02865v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.02865", "catalog_id": "arxiv:2506.02865", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-03", "identifiers": { "arxiv": "2506.02865" }, "record_updated_at": "2025-06-11" }, "genre": "preprint", "id": "arxiv:2506.02865", "issued": { "date-parts": [ [ 2025, 6, 3 ] ] }, "note": "First published: 2025-06-03; Recorded arXiv version: v2; Recorded version updated: 2025-06-11T09:33:38Z.", "title": "Surfer-H Meets Holo1: Cost-Efficient Web Agent Powered by Open Weights", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2506.03095v1", "archive": "arXiv", "archive_location": "2506.03095", "author": [ { "literal": "Man Luo" }, { "literal": "David Cobbley" }, { "literal": "Xin Su" }, { "literal": "Shachar Rosenman" }, { "literal": "Vasudev Lal" }, { "literal": "Shao-Yen Tseng" }, { "literal": "Phillip Howard" } ], "citation-key": "arxiv_2506_03095", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "068eae2cab9899967850f21616a701f3eeb258b9654c83d34f68be266e48854e", "submitted_at": "2025-06-03T17:27:04Z", "updated_at": "2025-06-03T17:27:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.03095v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.03095", "catalog_id": "arxiv:2506.03095", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-03", "identifiers": { "arxiv": "2506.03095" }, "record_updated_at": "2025-06-03" }, "genre": "preprint", "id": "arxiv:2506.03095", "issued": { "date-parts": [ [ 2025, 6, 3 ] ] }, "note": "First published: 2025-06-03; Recorded arXiv version: v1; Recorded version updated: 2025-06-03T17:27:04Z.", "title": "DPO Learning with LLMs-Judge Signal for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.04135v4", "archive": "arXiv", "archive_location": "2506.04135", "author": [ { "literal": "Pei Yang" }, { "literal": "Hai Ci" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2506_04135", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8a4c980694819dc5eb4ff322e6a703cccfe0b1e45efb60f22b120592d03edf36", "submitted_at": "2025-06-04T16:26:56Z", "updated_at": "2025-10-18T12:11:31Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2506.04135v4" }, "canonical_source_url": "https://arxiv.org/abs/2506.04135", "catalog_id": "arxiv:2506.04135", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-04", "identifiers": { "arxiv": "2506.04135" }, "record_updated_at": "2025-10-18" }, "genre": "preprint", "id": "arxiv:2506.04135", "issued": { "date-parts": [ [ 2025, 6, 4 ] ] }, "note": "First published: 2025-06-04; Recorded arXiv version: v4; Recorded version updated: 2025-10-18T12:11:31Z.", "title": "macOSWorld: A Multilingual Interactive Benchmark for GUI Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2506.07217v2", "archive": "arXiv", "archive_location": "2506.07217", "author": [ { "literal": "Zihan Deng" }, { "literal": "Changyu Du" }, { "literal": "Stavros Nousias" }, { "literal": "André Borrmann" } ], "citation-key": "arxiv_2506_07217", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2bf4f91de32fdf3bd9b81ba64ace02e4a7d02ff094794f30dfd560389f334cea", "submitted_at": "2025-06-08T16:45:31Z", "updated_at": "2025-06-30T08:31:07Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.07217v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.07217", "catalog_id": "arxiv:2506.07217", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-08", "identifiers": { "arxiv": "2506.07217" }, "record_updated_at": "2025-06-30" }, "genre": "preprint", "id": "arxiv:2506.07217", "issued": { "date-parts": [ [ 2025, 6, 8 ] ] }, "note": "First published: 2025-06-08; Recorded arXiv version: v2; Recorded version updated: 2025-06-30T08:31:07Z.", "title": "BIMgent: Towards Autonomous Building Modeling via Computer-use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2506.07385v1", "archive": "arXiv", "archive_location": "2506.07385", "author": [ { "literal": "Ruofan Liu" }, { "literal": "Xiwen Teoh" }, { "literal": "Yun Lin" }, { "literal": "Guanjie Chen" }, { "literal": "Ruofei Ren" }, { "literal": "Denys Poshyvanyk" }, { "literal": "Jin Song Dong" } ], "citation-key": "arxiv_2506_07385", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bea3e76b3124ebbcd8f7817b3e32d667d8aaae6e31f1b75ac3fc8c39e7637c6f", "submitted_at": "2025-06-09T03:09:48Z", "updated_at": "2025-06-09T03:09:48Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.07385v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.07385", "catalog_id": "arxiv:2506.07385", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-09", "identifiers": { "arxiv": "2506.07385" }, "record_updated_at": "2025-06-09" }, "genre": "preprint", "id": "arxiv:2506.07385", "issued": { "date-parts": [ [ 2025, 6, 9 ] ] }, "note": "First published: 2025-06-09; Recorded arXiv version: v1; Recorded version updated: 2025-06-09T03:09:48Z.", "title": "GUIPilot: A Consistency-based Mobile GUI Testing Approach for Detecting Application-specific Bugs", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.07672v1", "archive": "arXiv", "archive_location": "2506.07672", "author": [ { "literal": "Yunhe Yan" }, { "literal": "Shihe Wang" }, { "literal": "Jiajun Du" }, { "literal": "Yexuan Yang" }, { "literal": "Yuxuan Shan" }, { "literal": "Qichen Qiu" }, { "literal": "Xianqing Jia" }, { "literal": "Xinge Wang" }, { "literal": "Xin Yuan" }, { "literal": "Xu Han" }, { "literal": "Mao Qin" }, { "literal": "Yinxiao Chen" }, { "literal": "Chen Peng" }, { "literal": "Shangguang Wang" }, { "literal": "Mengwei Xu" } ], "citation-key": "arxiv_2506_07672", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2334e017fe49b39dc5d460963988f05d11051f2101bebe23be76005809fafd45", "submitted_at": "2025-06-09T11:50:33Z", "updated_at": "2025-06-09T11:50:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.07672v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.07672", "catalog_id": "arxiv:2506.07672", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-09", "identifiers": { "arxiv": "2506.07672" }, "record_updated_at": "2025-06-09" }, "genre": "preprint", "id": "arxiv:2506.07672", "issued": { "date-parts": [ [ 2025, 6, 9 ] ] }, "note": "First published: 2025-06-09; Recorded arXiv version: v1; Recorded version updated: 2025-06-09T11:50:33Z.", "title": "MCPWorld: A Unified Benchmarking Testbed for API, GUI, and Hybrid Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.08012v1", "archive": "arXiv", "archive_location": "2506.08012", "author": [ { "literal": "Penghao Wu" }, { "literal": "Shengnan Ma" }, { "literal": "Bo Wang" }, { "literal": "Jiaheng Yu" }, { "literal": "Lewei Lu" }, { "literal": "Ziwei Liu" } ], "citation-key": "arxiv_2506_08012", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ce3b7d2c25b5a99f110f00be7389d4230d625ac1f7f33d5ab81f3d6f441f3610", "submitted_at": "2025-06-09T17:59:57Z", "updated_at": "2025-06-09T17:59:57Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.08012v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.08012", "catalog_id": "arxiv:2506.08012", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-09", "identifiers": { "arxiv": "2506.08012" }, "record_updated_at": "2025-06-09" }, "genre": "preprint", "id": "arxiv:2506.08012", "issued": { "date-parts": [ [ 2025, 6, 9 ] ] }, "note": "First published: 2025-06-09; Recorded arXiv version: v1; Recorded version updated: 2025-06-09T17:59:57Z.", "title": "GUI-Reflection: Empowering Multimodal GUI Models with Self-Reflection Behavior", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.08136v3", "archive": "arXiv", "archive_location": "2506.08136", "author": [ { "literal": "Zefang Liu" }, { "literal": "Yinzhu Quan" } ], "citation-key": "arxiv_2506_08136", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "43aca3b0b8a1a0d5729a4e90dc1516bbb68e3611de3220876cdb29afc81d847f", "submitted_at": "2025-06-09T18:39:48Z", "updated_at": "2026-05-11T08:25:57Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2506.08136v3" }, "canonical_source_url": "https://arxiv.org/abs/2506.08136", "catalog_id": "arxiv:2506.08136", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-09", "identifiers": { "arxiv": "2506.08136" }, "record_updated_at": "2026-05-11" }, "genre": "preprint", "id": "arxiv:2506.08136", "issued": { "date-parts": [ [ 2025, 6, 9 ] ] }, "note": "First published: 2025-06-09; Recorded arXiv version: v3; Recorded version updated: 2026-05-11T08:25:57Z.", "title": "EconWebArena: Benchmarking Autonomous Agents on Economic Tasks in Realistic Web Environments", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2506.08972v1", "archive": "arXiv", "archive_location": "2506.08972", "author": [ { "literal": "Yuan Guo" }, { "literal": "Tingjia Miao" }, { "literal": "Zheng Wu" }, { "literal": "Pengzhou Cheng" }, { "literal": "Ming Zhou" }, { "literal": "Zhuosheng Zhang" } ], "citation-key": "arxiv_2506_08972", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8dbc84aa582b94775d490cf84e17eff05be84af7dbcc2de7de61b886fa3c9178", "submitted_at": "2025-06-10T16:45:29Z", "updated_at": "2025-06-10T16:45:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.08972v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.08972", "catalog_id": "arxiv:2506.08972", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-10", "identifiers": { "arxiv": "2506.08972" }, "record_updated_at": "2025-06-10" }, "genre": "preprint", "id": "arxiv:2506.08972", "issued": { "date-parts": [ [ 2025, 6, 10 ] ] }, "note": "First published: 2025-06-10; Recorded arXiv version: v1; Recorded version updated: 2025-06-10T16:45:29Z.", "title": "Atomic-to-Compositional Generalization for Mobile Agents with A New Benchmark and Scheduling System", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.14205v2", "archive": "arXiv", "archive_location": "2506.14205", "author": [ { "literal": "Jingxu Xie" }, { "literal": "Dylan Xu" }, { "literal": "Xuandong Zhao" }, { "literal": "Dawn Song" } ], "citation-key": "arxiv_2506_14205", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ae36b271f9b4799093831cabee772845edfaeca97107414e424e170ccf89904d", "submitted_at": "2025-06-17T05:46:52Z", "updated_at": "2026-03-02T00:51:36Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.14205v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.14205", "catalog_id": "arxiv:2506.14205", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-17", "identifiers": { "arxiv": "2506.14205" }, "record_updated_at": "2026-03-02" }, "genre": "preprint", "id": "arxiv:2506.14205", "issued": { "date-parts": [ [ 2025, 6, 17 ] ] }, "note": "First published: 2025-06-17; Recorded arXiv version: v2; Recorded version updated: 2026-03-02T00:51:36Z.", "title": "AgentSynth: Scalable Task Generation for Generalist Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2506.14477v1", "archive": "arXiv", "archive_location": "2506.14477", "author": [ { "literal": "Jingqi Yang" }, { "literal": "Zhilong Song" }, { "literal": "Jiawei Chen" }, { "literal": "Mingli Song" }, { "literal": "Sheng Zhou" }, { "literal": "linjun sun" }, { "literal": "Xiaogang Ouyang" }, { "literal": "Chun Chen" }, { "literal": "Can Wang" } ], "citation-key": "arxiv_2506_14477", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f9e64118f03c2a9f3f55e35f79c38835f1eb823265c83c933fdceca2610b9904", "submitted_at": "2025-06-17T12:50:35Z", "updated_at": "2025-06-17T12:50:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.14477v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.14477", "catalog_id": "arxiv:2506.14477", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-17", "identifiers": { "arxiv": "2506.14477" }, "record_updated_at": "2025-06-17" }, "genre": "preprint", "id": "arxiv:2506.14477", "issued": { "date-parts": [ [ 2025, 6, 17 ] ] }, "note": "First published: 2025-06-17; Recorded arXiv version: v1; Recorded version updated: 2025-06-17T12:50:35Z.", "title": "GUI-Robust: A Comprehensive Dataset for Testing GUI Agent Robustness in Real-World Anomalies", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.14866v2", "archive": "arXiv", "archive_location": "2506.14866", "author": [ { "literal": "Thomas Kuntz" }, { "literal": "Agatha Duzan" }, { "literal": "Hao Zhao" }, { "literal": "Francesco Croce" }, { "literal": "Zico Kolter" }, { "literal": "Nicolas Flammarion" }, { "literal": "Maksym Andriushchenko" } ], "citation-key": "arxiv_2506_14866", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f551b861012fc4eddbf010a619bc65e77b524b05958761b738a88bcae377e9a9", "submitted_at": "2025-06-17T17:59:31Z", "updated_at": "2025-10-29T10:34:04Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.14866v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.14866", "catalog_id": "arxiv:2506.14866", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-17", "identifiers": { "arxiv": "2506.14866" }, "record_updated_at": "2025-10-29" }, "genre": "preprint", "id": "arxiv:2506.14866", "issued": { "date-parts": [ [ 2025, 6, 17 ] ] }, "note": "First published: 2025-06-17; Recorded arXiv version: v2; Recorded version updated: 2025-10-29T10:34:04Z.", "title": "OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2506.15425v1", "archive": "arXiv", "archive_location": "2506.15425", "author": [ { "literal": "Xingjian Tao" }, { "literal": "Yiwei Wang" }, { "literal": "Yujun Cai" }, { "literal": "Zhicheng Yang" }, { "literal": "Jing Tang" } ], "citation-key": "arxiv_2506_15425", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e3884b1807cbee971364947aa2f3756937c4c774d9942bb4da3d0fabb69bda07", "submitted_at": "2025-06-18T12:55:35Z", "updated_at": "2025-06-18T12:55:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.15425v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.15425", "catalog_id": "arxiv:2506.15425", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-18", "identifiers": { "arxiv": "2506.15425" }, "record_updated_at": "2025-06-18" }, "genre": "preprint", "id": "arxiv:2506.15425", "issued": { "date-parts": [ [ 2025, 6, 18 ] ] }, "note": "First published: 2025-06-18; Recorded arXiv version: v1; Recorded version updated: 2025-06-18T12:55:35Z.", "title": "Understanding GUI Agent Localization Biases through Logit Sharpness", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.15677v3", "archive": "arXiv", "archive_location": "2506.15677", "author": [ { "literal": "Yining Hong" }, { "literal": "Rui Sun" }, { "literal": "Bingxuan Li" }, { "literal": "Xingcheng Yao" }, { "literal": "Maxine Wu" }, { "literal": "Alexander Chien" }, { "literal": "Da Yin" }, { "literal": "Ying Nian Wu" }, { "literal": "Zhecan James Wang" }, { "literal": "Kai-Wei Chang" } ], "citation-key": "arxiv_2506_15677", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3e8f714b234a1aad60d0306df23c90410aebfd257fb5906644d0382479e8f3d0", "submitted_at": "2025-06-18T17:58:17Z", "updated_at": "2025-07-29T22:40:49Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2506.15677v3" }, "canonical_source_url": "https://arxiv.org/abs/2506.15677", "catalog_id": "arxiv:2506.15677", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-18", "identifiers": { "arxiv": "2506.15677" }, "record_updated_at": "2025-07-29" }, "genre": "preprint", "id": "arxiv:2506.15677", "issued": { "date-parts": [ [ 2025, 6, 18 ] ] }, "note": "First published: 2025-06-18; Recorded arXiv version: v3; Recorded version updated: 2025-07-29T22:40:49Z.", "title": "Embodied Web Agents: Bridging Physical-Digital Realms for Integrated Agent Intelligence", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2506.16042v2", "archive": "arXiv", "archive_location": "2506.16042", "author": [ { "literal": "Reyna Abhyankar" }, { "literal": "Qi Qi" }, { "literal": "Yiying Zhang" } ], "citation-key": "arxiv_2506_16042", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "df3f654ef1364aeb6cdaeaea1c6334da949a798b2f8256c9ddc4e6a88a588e19", "submitted_at": "2025-06-19T05:26:40Z", "updated_at": "2026-05-18T17:55:59Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.16042v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.16042", "catalog_id": "arxiv:2506.16042", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-19", "identifiers": { "arxiv": "2506.16042" }, "record_updated_at": "2026-05-18" }, "genre": "preprint", "id": "arxiv:2506.16042", "issued": { "date-parts": [ [ 2025, 6, 19 ] ] }, "note": "First published: 2025-06-19; Recorded arXiv version: v2; Recorded version updated: 2026-05-18T17:55:59Z.", "title": "OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2506.17318v1", "archive": "arXiv", "archive_location": "2506.17318", "author": [ { "literal": "Atharv Singh Patlan" }, { "literal": "Ashwin Hebbar" }, { "literal": "Pramod Viswanath" }, { "literal": "Prateek Mittal" } ], "citation-key": "arxiv_2506_17318", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2d346389013d465f578417a7fca80ee527f5e7b582e93eb73dae148f11479971", "submitted_at": "2025-06-18T14:29:02Z", "updated_at": "2025-06-18T14:29:02Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2506.17318v1" }, "canonical_source_url": "https://arxiv.org/abs/2506.17318", "catalog_id": "arxiv:2506.17318", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-18", "identifiers": { "arxiv": "2506.17318" }, "record_updated_at": "2025-06-18" }, "genre": "preprint", "id": "arxiv:2506.17318", "issued": { "date-parts": [ [ 2025, 6, 18 ] ] }, "note": "First published: 2025-06-18; Recorded arXiv version: v1; Recorded version updated: 2025-06-18T14:29:02Z.", "title": "Context manipulation attacks : Web agents are susceptible to corrupted memory", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2506.17913v2", "archive": "arXiv", "archive_location": "2506.17913", "author": [ { "literal": "Jinjie Wei" }, { "literal": "Jiyao Liu" }, { "literal": "Lihao Liu" }, { "literal": "Ming Hu" }, { "literal": "Junzhi Ning" }, { "literal": "Mingcheng Li" }, { "literal": "Weijie Yin" }, { "literal": "Junjun He" }, { "literal": "Xiao Liang" }, { "literal": "Chao Feng" }, { "literal": "Dingkang Yang" } ], "citation-key": "arxiv_2506_17913", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f12d2ae58243b5f89a6a510e8f3ef4073a605c4bafdf63c37fd38db57ea84602", "submitted_at": "2025-06-22T06:30:52Z", "updated_at": "2026-07-21T05:30:48Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2506.17913v2" }, "canonical_source_url": "https://arxiv.org/abs/2506.17913", "catalog_id": "arxiv:2506.17913", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-22", "identifiers": { "arxiv": "2506.17913" }, "record_updated_at": "2026-07-21" }, "genre": "preprint", "id": "arxiv:2506.17913", "issued": { "date-parts": [ [ 2025, 6, 22 ] ] }, "note": "First published: 2025-06-22; Recorded arXiv version: v2; Recorded version updated: 2026-07-21T05:30:48Z.", "title": "Learning, Reasoning, Refinement: A Framework for Kahneman's Dual-System Intelligence in GUI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2507.00841v1", "archive": "arXiv", "archive_location": "2507.00841", "author": [ { "literal": "Siyuan Liang" }, { "literal": "Tianmeng Fang" }, { "literal": "Zhe Liu" }, { "literal": "Aishan Liu" }, { "literal": "Yan Xiao" }, { "literal": "Jinyuan He" }, { "literal": "Ee-Chien Chang" }, { "literal": "Xiaochun Cao" } ], "citation-key": "arxiv_2507_00841", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "78dd523ad6123fe1ee5164685a3853719c9a4442adac8661050f147ff6a9f619", "submitted_at": "2025-07-01T15:10:00Z", "updated_at": "2025-07-01T15:10:00Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2507.00841v1" }, "canonical_source_url": "https://arxiv.org/abs/2507.00841", "catalog_id": "arxiv:2507.00841", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-01", "identifiers": { "arxiv": "2507.00841" }, "record_updated_at": "2025-07-01" }, "genre": "preprint", "id": "arxiv:2507.00841", "issued": { "date-parts": [ [ 2025, 7, 1 ] ] }, "note": "First published: 2025-07-01; Recorded arXiv version: v1; Recorded version updated: 2025-07-01T15:10:00Z.", "title": "SafeMobile: Chain-level Jailbreak Detection and Automated Evaluation for Multimodal Mobile Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2507.00938v4", "archive": "arXiv", "archive_location": "2507.00938", "author": [ { "literal": "Zihao Sun" }, { "literal": "Zijing Shi" }, { "literal": "Ling Chen" } ], "citation-key": "arxiv_2507_00938", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1745ec9cf10c465c0f389f885df6db502f1884cca271bccea1b45acd2c787af9", "submitted_at": "2025-07-01T16:43:57Z", "updated_at": "2026-09-24T12:58:22Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2507.00938v4" }, "canonical_source_url": "https://arxiv.org/abs/2507.00938", "catalog_id": "arxiv:2507.00938", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-01", "identifiers": { "arxiv": "2507.00938" }, "record_updated_at": "2026-09-24" }, "genre": "preprint", "id": "arxiv:2507.00938", "issued": { "date-parts": [ [ 2025, 7, 1 ] ] }, "note": "First published: 2025-07-01; Recorded arXiv version: v4; Recorded version updated: 2026-09-24T12:58:22Z.", "title": "WebArxiv: A Reproducible Benchmark for Evaluating Multimodal Web Agents on arXiv Tasks", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2507.04103v4", "archive": "arXiv", "archive_location": "2507.04103", "author": [ { "literal": "Dheeraj Vattikonda" }, { "literal": "Santhoshi Ravichandran" }, { "literal": "Emiliano Penaloza" }, { "literal": "Hadi Nekoei" }, { "literal": "Megh Thakkar" }, { "literal": "Thibault Le Sellier de Chezelles" }, { "literal": "Nicolas Gontier" }, { "literal": "Miguel Muñoz-Mármol" }, { "literal": "Sahar Omidi Shayegan" }, { "literal": "Stefania Raimondo" }, { "literal": "Xue Liu" }, { "literal": "Alexandre Drouin" }, { "literal": "Laurent Charlin" }, { "literal": "Alexandre Piché" }, { "literal": "Alexandre Lacoste" }, { "literal": "Massimo Caccia" } ], "citation-key": "arxiv_2507_04103", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "15fa527a869ab66f4f39900ab07ea8b8db930703a8e950ed1e195e1e66164099", "submitted_at": "2025-07-05T17:12:33Z", "updated_at": "2026-02-13T17:24:17Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2507.04103v4" }, "canonical_source_url": "https://arxiv.org/abs/2507.04103", "catalog_id": "arxiv:2507.04103", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-05", "identifiers": { "arxiv": "2507.04103" }, "record_updated_at": "2026-02-13" }, "genre": "preprint", "id": "arxiv:2507.04103", "issued": { "date-parts": [ [ 2025, 7, 5 ] ] }, "note": "First published: 2025-07-05; Recorded arXiv version: v4; Recorded version updated: 2026-02-13T17:24:17Z.", "title": "How to Train Your LLM Web Agent: A Statistical Diagnosis", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2507.04227v2", "archive": "arXiv", "archive_location": "2507.04227", "author": [ { "literal": "Guohong Liu" }, { "literal": "Jialei Ye" }, { "literal": "Jiacheng Liu" }, { "literal": "Yuanchun Li" }, { "literal": "Wei Liu" }, { "literal": "Pengzhi Gao" }, { "literal": "Jian Luan" }, { "literal": "Yunxin Liu" } ], "citation-key": "arxiv_2507_04227", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "85ebf8b3b5916e66e4b4221c73bb29519156c41e8d1695bcc46d275dd50cf58d", "submitted_at": "2025-07-06T03:31:36Z", "updated_at": "2026-04-14T14:47:25Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2507.04227v2" }, "canonical_source_url": "https://arxiv.org/abs/2507.04227", "catalog_id": "arxiv:2507.04227", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-06", "identifiers": { "arxiv": "2507.04227" }, "record_updated_at": "2026-04-14" }, "genre": "preprint", "id": "arxiv:2507.04227", "issued": { "date-parts": [ [ 2025, 7, 6 ] ] }, "note": "First published: 2025-07-06; Recorded arXiv version: v2; Recorded version updated: 2026-04-14T14:47:25Z.", "title": "Mobile GUI Agents under Real-world Threats: Are We There Yet?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2507.05445v1", "archive": "arXiv", "archive_location": "2507.05445", "author": [ { "literal": "Daniel Jones" }, { "literal": "Giorgio Severi" }, { "literal": "Martin Pouliot" }, { "literal": "Gary Lopez" }, { "literal": "Joris de Gruyter" }, { "literal": "Santiago Zanella-Beguelin" }, { "literal": "Justin Song" }, { "literal": "Blake Bullwinkel" }, { "literal": "Pamela Cortez" }, { "literal": "Amanda Minnich" } ], "citation-key": "arxiv_2507_05445", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3ad2f299303076e3474cf247cf25d20ba63b1f1aa8601ab74243c00896bb56e4", "submitted_at": "2025-07-07T19:50:21Z", "updated_at": "2025-07-07T19:50:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2507.05445v1" }, "canonical_source_url": "https://arxiv.org/abs/2507.05445", "catalog_id": "arxiv:2507.05445", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-07", "identifiers": { "arxiv": "2507.05445" }, "record_updated_at": "2025-07-07" }, "genre": "preprint", "id": "arxiv:2507.05445", "issued": { "date-parts": [ [ 2025, 7, 7 ] ] }, "note": "First published: 2025-07-07; Recorded arXiv version: v1; Recorded version updated: 2025-07-07T19:50:21Z.", "title": "A Systematization of Security Vulnerabilities in Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2507.06134v2", "archive": "arXiv", "archive_location": "2507.06134", "author": [ { "literal": "Sanidhya Vijayvargiya" }, { "literal": "Aditya Bharat Soni" }, { "literal": "Xuhui Zhou" }, { "literal": "Zora Zhiruo Wang" }, { "literal": "Nouha Dziri" }, { "literal": "Graham Neubig" }, { "literal": "Maarten Sap" } ], "citation-key": "arxiv_2507_06134", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "db56d120a7f04f9181a107bcd75c61a3eb3ac36939e2a0517d622332c70f3eac", "submitted_at": "2025-07-08T16:18:54Z", "updated_at": "2026-02-16T22:34:53Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2507.06134v2" }, "canonical_source_url": "https://arxiv.org/abs/2507.06134", "catalog_id": "arxiv:2507.06134", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-08", "identifiers": { "arxiv": "2507.06134" }, "record_updated_at": "2026-02-16" }, "genre": "preprint", "id": "arxiv:2507.06134", "issued": { "date-parts": [ [ 2025, 7, 8 ] ] }, "note": "First published: 2025-07-08; Recorded arXiv version: v2; Recorded version updated: 2026-02-16T22:34:53Z.", "title": "OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2507.06899v2", "archive": "arXiv", "archive_location": "2507.06899", "author": [ { "literal": "Ziang Ye" }, { "literal": "Yang Zhang" }, { "literal": "Wentao Shi" }, { "literal": "Xiaoyu You" }, { "literal": "Fuli Feng" }, { "literal": "Tat-Seng Chua" } ], "citation-key": "arxiv_2507_06899", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "010a3a54ba37417c97e52434211375447418dd2e7ce1da6c339532534f0a920e", "submitted_at": "2025-07-09T14:36:00Z", "updated_at": "2025-09-24T14:33:17Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2507.06899v2" }, "canonical_source_url": "https://arxiv.org/abs/2507.06899", "catalog_id": "arxiv:2507.06899", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-09", "identifiers": { "arxiv": "2507.06899" }, "record_updated_at": "2025-09-24" }, "genre": "preprint", "id": "arxiv:2507.06899", "issued": { "date-parts": [ [ 2025, 7, 9 ] ] }, "note": "First published: 2025-07-09; Recorded arXiv version: v2; Recorded version updated: 2025-09-24T14:33:17Z.", "title": "VisualTrap: A Stealthy Backdoor Attack on GUI Agents via Visual Grounding Manipulation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2507.12844v2", "archive": "arXiv", "archive_location": "2507.12844", "author": [ { "literal": "Joel Nitu" }, { "literal": "Heidrun Mühle" }, { "literal": "Andreas Stöckl" } ], "citation-key": "arxiv_2507_12844", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9fe2d9d877a0666d46a065bd8b673c282c879ef0a2620e3b267793a91858b519", "submitted_at": "2025-07-17T07:10:27Z", "updated_at": "2025-11-12T13:34:24Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2507.12844v2" }, "canonical_source_url": "https://arxiv.org/abs/2507.12844", "catalog_id": "arxiv:2507.12844", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-17", "identifiers": { "arxiv": "2507.12844" }, "record_updated_at": "2025-11-12" }, "genre": "preprint", "id": "arxiv:2507.12844", "issued": { "date-parts": [ [ 2025, 7, 17 ] ] }, "note": "First published: 2025-07-17; Recorded arXiv version: v2; Recorded version updated: 2025-11-12T13:34:24Z.", "title": "Machine-Readable Ads: Accessibility and Trust Patterns for AI Web Agents interacting with Online Advertisements", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2507.14293v1", "archive": "arXiv", "archive_location": "2507.14293", "author": [ { "literal": "Boyuan Zheng" }, { "literal": "Zeyi Liao" }, { "literal": "Scott Salisbury" }, { "literal": "Zeyuan Liu" }, { "literal": "Michael Lin" }, { "literal": "Qinyuan Zheng" }, { "literal": "Zifan Wang" }, { "literal": "Xiang Deng" }, { "literal": "Dawn Song" }, { "literal": "Huan Sun" }, { "literal": "Yu Su" } ], "citation-key": "arxiv_2507_14293", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a8f082fd40174e01f40e97d64f7aa3d90e5f76e831df8f93afdc21aa2e35bfd5", "submitted_at": "2025-07-18T18:06:27Z", "updated_at": "2025-07-18T18:06:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2507.14293v1" }, "canonical_source_url": "https://arxiv.org/abs/2507.14293", "catalog_id": "arxiv:2507.14293", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-18", "identifiers": { "arxiv": "2507.14293" }, "record_updated_at": "2025-07-18" }, "genre": "preprint", "id": "arxiv:2507.14293", "issued": { "date-parts": [ [ 2025, 7, 18 ] ] }, "note": "First published: 2025-07-18; Recorded arXiv version: v1; Recorded version updated: 2025-07-18T18:06:27Z.", "title": "WebGuard: Building a Generalizable Guardrail for Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2507.14799v1", "archive": "arXiv", "archive_location": "2507.14799", "author": [ { "literal": "Sam Johnson" }, { "literal": "Viet Pham" }, { "literal": "Thai Le" } ], "citation-key": "arxiv_2507_14799", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "79963ac401384f0c23ecc65d58e616daacc93103456ceb39241dcd6d71c6591c", "submitted_at": "2025-07-20T03:10:13Z", "updated_at": "2025-07-20T03:10:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2507.14799v1" }, "canonical_source_url": "https://arxiv.org/abs/2507.14799", "catalog_id": "arxiv:2507.14799", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-20", "identifiers": { "arxiv": "2507.14799" }, "record_updated_at": "2025-07-20" }, "genre": "preprint", "id": "arxiv:2507.14799", "issued": { "date-parts": [ [ 2025, 7, 20 ] ] }, "note": "First published: 2025-07-20; Recorded arXiv version: v1; Recorded version updated: 2025-07-20T03:10:13Z.", "title": "Manipulating LLM Web Agents with Indirect Prompt Injection Attack via HTML Accessibility Tree", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2507.19132v1", "archive": "arXiv", "archive_location": "2507.19132", "author": [ { "literal": "Xuetian Chen" }, { "literal": "Yinghao Chen" }, { "literal": "Xinfeng Yuan" }, { "literal": "Zhuo Peng" }, { "literal": "Lu Chen" }, { "literal": "Yuekeng Li" }, { "literal": "Zhoujia Zhang" }, { "literal": "Yingqian Huang" }, { "literal": "Leyan Huang" }, { "literal": "Jiaqing Liang" }, { "literal": "Tianbao Xie" }, { "literal": "Zhiyong Wu" }, { "literal": "Qiushi Sun" }, { "literal": "Biqing Qi" }, { "literal": "Bowen Zhou" } ], "citation-key": "arxiv_2507_19132", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fcd5a34bf6187f48da8e08842678e1e21cd558b2bcc7aab2611ceee7bc0d0179", "submitted_at": "2025-07-25T10:14:53Z", "updated_at": "2025-07-25T10:14:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2507.19132v1" }, "canonical_source_url": "https://arxiv.org/abs/2507.19132", "catalog_id": "arxiv:2507.19132", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-25", "identifiers": { "arxiv": "2507.19132" }, "record_updated_at": "2025-07-25" }, "genre": "preprint", "id": "arxiv:2507.19132", "issued": { "date-parts": [ [ 2025, 7, 25 ] ] }, "note": "First published: 2025-07-25; Recorded arXiv version: v1; Recorded version updated: 2025-07-25T10:14:53Z.", "title": "OS-MAP: How Far Can Computer-Using Agents Go in Breadth and Depth?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2507.19478v1", "archive": "arXiv", "archive_location": "2507.19478", "author": [ { "literal": "Xuehui Wang" }, { "literal": "Zhenyu Wu" }, { "literal": "JingJing Xie" }, { "literal": "Zichen Ding" }, { "literal": "Bowen Yang" }, { "literal": "Zehao Li" }, { "literal": "Zhaoyang Liu" }, { "literal": "Qingyun Li" }, { "literal": "Xuan Dong" }, { "literal": "Zhe Chen" }, { "literal": "Weiyun Wang" }, { "literal": "Xiangyu Zhao" }, { "literal": "Jixuan Chen" }, { "literal": "Haodong Duan" }, { "literal": "Tianbao Xie" }, { "literal": "Chenyu Yang" }, { "literal": "Shiqian Su" }, { "literal": "Yue Yu" }, { "literal": "Yuan Huang" }, { "literal": "Yiqian Liu" }, { "literal": "Xiao Zhang" }, { "literal": "Yanting Zhang" }, { "literal": "Xiangyu Yue" }, { "literal": "Weijie Su" }, { "literal": "Xizhou Zhu" }, { "literal": "Wei Shen" }, { "literal": "Jifeng Dai" }, { "literal": "Wenhai Wang" } ], "citation-key": "arxiv_2507_19478", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8b91f7bacbac917e374aa6f656dcfbcb2e220ab3227fa2cff295c63a1c005f9f", "submitted_at": "2025-07-25T17:59:26Z", "updated_at": "2025-07-25T17:59:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2507.19478v1" }, "canonical_source_url": "https://arxiv.org/abs/2507.19478", "catalog_id": "arxiv:2507.19478", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-07-25", "identifiers": { "arxiv": "2507.19478" }, "record_updated_at": "2025-07-25" }, "genre": "preprint", "id": "arxiv:2507.19478", "issued": { "date-parts": [ [ 2025, 7, 25 ] ] }, "note": "First published: 2025-07-25; Recorded arXiv version: v1; Recorded version updated: 2025-07-25T17:59:26Z.", "title": "MMBench-GUI: Hierarchical Multi-Platform Evaluation Framework for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2507.21071v2", "archive": "arXiv", "archive_location": "2507.21071", "author": [ { "literal": "Qinglong Yang" }, { "literal": "Haoming Li" }, { "literal": "Haotian Zhao" }, { "literal": "Xiaokai Yan" }, { "literal": "Jingtao Ding" }, { "literal": "Fengli Xu" }, { "literal": "Yong Li" } ], "citation-key": "arxiv_2507_21071", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "947c85efc189635a118b823b8ca0787ff6b1c922d1af8b21a8686b7df9a27cf8", "submitted_at": "2025-06-09T06:38:41Z", "updated_at": "2026-03-16T09:14:01Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2507.21071v2" }, "canonical_source_url": "https://arxiv.org/abs/2507.21071", "catalog_id": "arxiv:2507.21071", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-06-09", "identifiers": { "arxiv": "2507.21071" }, "record_updated_at": "2026-03-16" }, "genre": "preprint", "id": "arxiv:2507.21071", "issued": { "date-parts": [ [ 2025, 6, 9 ] ] }, "note": "First published: 2025-06-09; Recorded arXiv version: v2; Recorded version updated: 2026-03-16T09:14:01Z.", "title": "FingerTip 20K: A Benchmark for Proactive and Personalized Mobile LLM Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2508.01330v4", "archive": "arXiv", "archive_location": "2508.01330", "author": [ { "literal": "Zihan Zheng" }, { "literal": "Tianle Cui" }, { "literal": "Taoran Wang" }, { "literal": "Fengtao Wang" }, { "literal": "Jiahui Pan" }, { "literal": "Lewei He" }, { "literal": "Qianglong Chen" } ], "citation-key": "arxiv_2508_01330", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "57d89909534af90cbc10b97a992238bde3ae976ac6cc97778b4345635cef552c", "submitted_at": "2025-08-02T11:53:41Z", "updated_at": "2026-04-20T02:44:32Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2508.01330v4" }, "canonical_source_url": "https://arxiv.org/abs/2508.01330", "catalog_id": "arxiv:2508.01330", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-02", "identifiers": { "arxiv": "2508.01330" }, "record_updated_at": "2026-04-20" }, "genre": "preprint", "id": "arxiv:2508.01330", "issued": { "date-parts": [ [ 2025, 8, 2 ] ] }, "note": "First published: 2025-08-02; Recorded arXiv version: v4; Recorded version updated: 2026-04-20T02:44:32Z.", "title": "NaturalGAIA: A Verifiable Benchmark and Hierarchical Framework for Long-Horizon GUI Tasks", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2508.01858v3", "archive": "arXiv", "archive_location": "2508.01858", "author": [ { "literal": "Yuhan Guo" }, { "literal": "Cong Guo" }, { "literal": "Aiwen Sun" }, { "literal": "Hongliang He" }, { "literal": "Xinyu Yang" }, { "literal": "Yue Lu" }, { "literal": "Yingji Zhang" }, { "literal": "Xuntao Guo" }, { "literal": "Dong Zhang" }, { "literal": "Jianzhuang Liu" }, { "literal": "Jiang Duan" }, { "literal": "Yijia Xiao" }, { "literal": "Liangjian Wen" }, { "literal": "Hai-Ming Xu" }, { "literal": "Yong Dai" } ], "citation-key": "arxiv_2508_01858", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7406b48ed9826d51215d955c31c3c67cd2ce9e08bf9a93a55ac4e85a546315f1", "submitted_at": "2025-08-03T17:17:52Z", "updated_at": "2026-07-05T06:17:25Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2508.01858v3" }, "canonical_source_url": "https://arxiv.org/abs/2508.01858", "catalog_id": "arxiv:2508.01858", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-03", "identifiers": { "arxiv": "2508.01858" }, "record_updated_at": "2026-07-05" }, "genre": "preprint", "id": "arxiv:2508.01858", "issued": { "date-parts": [ [ 2025, 8, 3 ] ] }, "note": "First published: 2025-08-03; Recorded arXiv version: v3; Recorded version updated: 2026-07-05T06:17:25Z.", "title": "Web-CogReasoner: Towards Multimodal Knowledge-Induced Cognitive Reasoning for Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2508.04025v1", "archive": "arXiv", "archive_location": "2508.04025", "author": [ { "literal": "Chao Hao" }, { "literal": "Shuai Wang" }, { "literal": "Kaiwen Zhou" } ], "citation-key": "arxiv_2508_04025", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "499a462f9f14b8afdd367bb7cf7a7ffb1b2e166c212fb5c36686d3145f08ab5f", "submitted_at": "2025-08-06T02:38:02Z", "updated_at": "2025-08-06T02:38:02Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2508.04025v1" }, "canonical_source_url": "https://arxiv.org/abs/2508.04025", "catalog_id": "arxiv:2508.04025", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-06", "identifiers": { "arxiv": "2508.04025" }, "record_updated_at": "2025-08-06" }, "genre": "preprint", "id": "arxiv:2508.04025", "issued": { "date-parts": [ [ 2025, 8, 6 ] ] }, "note": "First published: 2025-08-06; Recorded arXiv version: v1; Recorded version updated: 2025-08-06T02:38:02Z.", "title": "Uncertainty-Aware GUI Agent: Adaptive Perception through Component Recommendation and Human-in-the-Loop Refinement", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2508.09057v2", "archive": "arXiv", "archive_location": "2508.09057", "author": [ { "literal": "Zeyu Huang" }, { "literal": "Juyuan Wang" }, { "literal": "Longfeng Chen" }, { "literal": "Boyi Xiao" }, { "literal": "Leng Cai" }, { "literal": "Yawen Zeng" }, { "literal": "Jin Xu" } ], "citation-key": "arxiv_2508_09057", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d4efa329c3b106219c9b25f675be0c561743085ac999d2964458a5f22ff6e62f", "submitted_at": "2025-08-12T16:18:30Z", "updated_at": "2025-08-14T16:36:45Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2508.09057v2" }, "canonical_source_url": "https://arxiv.org/abs/2508.09057", "catalog_id": "arxiv:2508.09057", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-12", "identifiers": { "arxiv": "2508.09057" }, "record_updated_at": "2025-08-14" }, "genre": "preprint", "id": "arxiv:2508.09057", "issued": { "date-parts": [ [ 2025, 8, 12 ] ] }, "note": "First published: 2025-08-12; Recorded arXiv version: v2; Recorded version updated: 2025-08-14T16:36:45Z.", "title": "MVISU-Bench: Benchmarking Mobile Agents for Real-World Tasks by Multi-App, Vague, Interactive, Single-App and Unethical Instructions", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2508.09241v1", "archive": "arXiv", "archive_location": "2508.09241", "author": [ { "literal": "Fengxian Ji" }, { "literal": "Jingpu Yang" }, { "literal": "Zirui Song" }, { "literal": "Yuanxi Wang" }, { "literal": "Zhexuan Cui" }, { "literal": "Yuke Li" }, { "literal": "Qian Jiang" }, { "literal": "Miao Fang" }, { "literal": "Xiuying Chen" } ], "citation-key": "arxiv_2508_09241", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f8c91a1f85602760c706e56426c0d2d870bbc101ca500aa3c7c391938c3f1676", "submitted_at": "2025-08-12T15:12:42Z", "updated_at": "2025-08-12T15:12:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2508.09241v1" }, "canonical_source_url": "https://arxiv.org/abs/2508.09241", "catalog_id": "arxiv:2508.09241", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-12", "identifiers": { "arxiv": "2508.09241" }, "record_updated_at": "2025-08-12" }, "genre": "preprint", "id": "arxiv:2508.09241", "issued": { "date-parts": [ [ 2025, 8, 12 ] ] }, "note": "First published: 2025-08-12; Recorded arXiv version: v1; Recorded version updated: 2025-08-12T15:12:42Z.", "title": "FineState-Bench: A Comprehensive Benchmark for Fine-Grained State Control in GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2508.13024v3", "archive": "arXiv", "archive_location": "2508.13024", "author": [ { "literal": "Ralph Peeters" }, { "literal": "Aaron Steiner" }, { "literal": "Luca Schwarz" }, { "literal": "Julian Yuya Caspary" }, { "literal": "Christian Bizer" } ], "citation-key": "arxiv_2508_13024", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0307d242a90f0650b5d5fdfbaf9653842d3ce1461d2653b0b1ba6b21c31ee2dc", "submitted_at": "2025-08-18T15:41:22Z", "updated_at": "2026-04-30T09:17:11Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2508.13024v3" }, "canonical_source_url": "https://arxiv.org/abs/2508.13024", "catalog_id": "arxiv:2508.13024", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-18", "identifiers": { "arxiv": "2508.13024" }, "record_updated_at": "2026-04-30" }, "genre": "preprint", "id": "arxiv:2508.13024", "issued": { "date-parts": [ [ 2025, 8, 18 ] ] }, "note": "First published: 2025-08-18; Recorded arXiv version: v3; Recorded version updated: 2026-04-30T09:17:11Z.", "title": "WebMall -- A Multi-Shop Benchmark for Evaluating Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2508.14104v1", "archive": "arXiv", "archive_location": "2508.14104", "author": [ { "literal": "Yutong Bian" }, { "literal": "Xianhao Lin" }, { "literal": "Yupeng Xie" }, { "literal": "Tianyang Liu" }, { "literal": "Mingchen Zhuge" }, { "literal": "Siyuan Lu" }, { "literal": "Haoming Tang" }, { "literal": "Jinlin Wang" }, { "literal": "Jiayi Zhang" }, { "literal": "Jiaqi Chen" }, { "literal": "Xiangru Tang" }, { "literal": "Yongxin Ni" }, { "literal": "Sirui Hong" }, { "literal": "Chenglin Wu" } ], "citation-key": "arxiv_2508_14104", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c5034c06e63607a1279668964e91c14ed762a95b6e3953337ef8b8b63c4fd57d", "submitted_at": "2025-08-17T07:31:11Z", "updated_at": "2025-08-17T07:31:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2508.14104v1" }, "canonical_source_url": "https://arxiv.org/abs/2508.14104", "catalog_id": "arxiv:2508.14104", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-17", "identifiers": { "arxiv": "2508.14104" }, "record_updated_at": "2025-08-17" }, "genre": "preprint", "id": "arxiv:2508.14104", "issued": { "date-parts": [ [ 2025, 8, 17 ] ] }, "note": "First published: 2025-08-17; Recorded arXiv version: v1; Recorded version updated: 2025-08-17T07:31:11Z.", "title": "You Don't Know Until You Click:Automated GUI Testing for Production-Ready Software Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2508.15832v2", "archive": "arXiv", "archive_location": "2508.15832", "author": [ { "literal": "Xianren Zhang" }, { "literal": "Shreyas Prasad" }, { "literal": "Di Wang" }, { "literal": "Qiuhai Zeng" }, { "literal": "Suhang Wang" }, { "literal": "Wenbo Yan" }, { "literal": "Mat Hans" } ], "citation-key": "arxiv_2508_15832", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6c8336a707723227a9d651a3e9347e45907f28ed7421b577761ed57f7d70593d", "submitted_at": "2025-08-18T21:58:43Z", "updated_at": "2026-04-21T00:38:54Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2508.15832v2" }, "canonical_source_url": "https://arxiv.org/abs/2508.15832", "catalog_id": "arxiv:2508.15832", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-18", "identifiers": { "arxiv": "2508.15832" }, "record_updated_at": "2026-04-21" }, "genre": "preprint", "id": "arxiv:2508.15832", "issued": { "date-parts": [ [ 2025, 8, 18 ] ] }, "note": "First published: 2025-08-18; Recorded arXiv version: v2; Recorded version updated: 2026-04-21T00:38:54Z.", "title": "A Functionality-Grounded Benchmark for Evaluating Web Agents in E-commerce Domains", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2508.19493v2", "archive": "arXiv", "archive_location": "2508.19493", "author": [ { "literal": "Zhixin Lin" }, { "literal": "Jungang Li" }, { "literal": "Shidong Pan" }, { "literal": "Yibo Shi" }, { "literal": "Yue Yao" }, { "literal": "Dongliang Xu" } ], "citation-key": "arxiv_2508_19493", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "815c8c1fa5993d2ef58798abd43a2b8945c476c27c777d30c1cbfacadd87af84", "submitted_at": "2025-08-27T00:41:28Z", "updated_at": "2025-09-03T07:48:34Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2508.19493v2" }, "canonical_source_url": "https://arxiv.org/abs/2508.19493", "catalog_id": "arxiv:2508.19493", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-27", "identifiers": { "arxiv": "2508.19493" }, "record_updated_at": "2025-09-03" }, "genre": "preprint", "id": "arxiv:2508.19493", "issued": { "date-parts": [ [ 2025, 8, 27 ] ] }, "note": "First published: 2025-08-27; Recorded arXiv version: v2; Recorded version updated: 2025-09-03T07:48:34Z.", "title": "Mind the Third Eye! Benchmarking Privacy Awareness in MLLM-powered Smartphone Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2508.19679v2", "archive": "arXiv", "archive_location": "2508.19679", "author": [ { "literal": "Qihang Ai" }, { "literal": "Pi Bu" }, { "literal": "Yue Cao" }, { "literal": "Yingyao Wang" }, { "literal": "Jihao Gu" }, { "literal": "Jingxuan Xing" }, { "literal": "Zekun Zhu" }, { "literal": "Wei Jiang" }, { "literal": "Zhicheng Zheng" }, { "literal": "Jun Song" }, { "literal": "Yuning Jiang" } ], "citation-key": "arxiv_2508_19679", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "385ed1b6e60c27660f0cbf281b0f2bbde052691801f7faae99cc053dfbc0d127", "submitted_at": "2025-08-27T08:40:05Z", "updated_at": "2026-04-25T13:29:08Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2508.19679v2" }, "canonical_source_url": "https://arxiv.org/abs/2508.19679", "catalog_id": "arxiv:2508.19679", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-27", "identifiers": { "arxiv": "2508.19679" }, "record_updated_at": "2026-04-25" }, "genre": "preprint", "id": "arxiv:2508.19679", "issued": { "date-parts": [ [ 2025, 8, 27 ] ] }, "note": "First published: 2025-08-27; Recorded arXiv version: v2; Recorded version updated: 2026-04-25T13:29:08Z.", "title": "InquireMobile: Teaching VLM-based Mobile Agent to Request Human Assistance via Reinforcement Fine-Tuning", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2508.21456v1", "archive": "arXiv", "archive_location": "2508.21456", "author": [ { "literal": "Yi-Hao Peng" }, { "literal": "Dingzeyu Li" }, { "literal": "Jeffrey P. Bigham" }, { "literal": "Amy Pavel" } ], "citation-key": "arxiv_2508_21456", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5dc6411807fd58daed4824e86c39377b24d9fd022ac19daceb836bfb382353cf", "submitted_at": "2025-08-29T09:39:00Z", "updated_at": "2025-08-29T09:39:00Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2508.21456v1" }, "canonical_source_url": "https://arxiv.org/abs/2508.21456", "catalog_id": "arxiv:2508.21456", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-29", "identifiers": { "arxiv": "2508.21456" }, "record_updated_at": "2025-08-29" }, "genre": "preprint", "id": "arxiv:2508.21456", "issued": { "date-parts": [ [ 2025, 8, 29 ] ] }, "note": "First published: 2025-08-29; Recorded arXiv version: v1; Recorded version updated: 2025-08-29T09:39:00Z.", "title": "Morae: Proactively Pausing UI Agents for User Choices", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.00366v1", "archive": "arXiv", "archive_location": "2509.00366", "author": [ { "literal": "Ziyi Guan" }, { "literal": "Jason Chun Lok Li" }, { "literal": "Zhijian Hou" }, { "literal": "Pingping Zhang" }, { "literal": "Donglai Xu" }, { "literal": "Yuzhi Zhao" }, { "literal": "Mengyang Wu" }, { "literal": "Jinpeng Chen" }, { "literal": "Thanh-Toan Nguyen" }, { "literal": "Pengfei Xian" }, { "literal": "Wenao Ma" }, { "literal": "Shengchao Qin" }, { "literal": "Graziano Chesi" }, { "literal": "Ngai Wong" } ], "citation-key": "arxiv_2509_00366", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c22f63bb5de22dd922ba37ea45df9404def017fee492e8ade03c83ab8b4f9b58", "submitted_at": "2025-08-30T05:32:32Z", "updated_at": "2025-08-30T05:32:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.00366v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.00366", "catalog_id": "arxiv:2509.00366", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-08-30", "identifiers": { "arxiv": "2509.00366" }, "record_updated_at": "2025-08-30" }, "genre": "preprint", "id": "arxiv:2509.00366", "issued": { "date-parts": [ [ 2025, 8, 30 ] ] }, "note": "First published: 2025-08-30; Recorded arXiv version: v1; Recorded version updated: 2025-08-30T05:32:32Z.", "title": "KG-RAG: Enhancing GUI Agent Decision-Making via Knowledge Graph-Driven Retrieval-Augmented Generation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.01052v2", "archive": "arXiv", "archive_location": "2509.01052", "author": [ { "literal": "Jaewoo Ahn" }, { "literal": "Junseo Kim" }, { "literal": "Heeseung Yun" }, { "literal": "Jaehyeon Son" }, { "literal": "Dongmin Park" }, { "literal": "Jaewoong Cho" }, { "literal": "Gunhee Kim" } ], "citation-key": "arxiv_2509_01052", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3bf86652aaeb252972f3010be38b75b78dfaa9a8063c480c80d0cea47397c2e7", "submitted_at": "2025-09-01T01:33:16Z", "updated_at": "2025-10-15T10:33:27Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2509.01052v2" }, "canonical_source_url": "https://arxiv.org/abs/2509.01052", "catalog_id": "arxiv:2509.01052", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2025-09-01", "identifiers": { "arxiv": "2509.01052" }, "record_updated_at": "2025-10-15" }, "genre": "preprint", "id": "arxiv:2509.01052", "issued": { "date-parts": [ [ 2025, 9, 1 ] ] }, "note": "First published: 2025-09-01; Recorded arXiv version: v2; Recorded version updated: 2025-10-15T10:33:27Z.", "title": "FlashAdventure: A Benchmark for GUI Agents Solving Full Story Arcs in Diverse Adventure Games", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2509.01619v1", "archive": "arXiv", "archive_location": "2509.01619", "author": [ { "literal": "Abhinav Kumar" }, { "literal": "Jaechul Roh" }, { "literal": "Ali Naseh" }, { "literal": "Amir Houmansadr" }, { "literal": "Eugene Bagdasarian" } ], "citation-key": "arxiv_2509_01619", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0f6ff68c59cede9cf59c039416a87ea39b926e3a3179b9f81af9354767befb93", "submitted_at": "2025-09-01T16:56:16Z", "updated_at": "2025-09-01T16:56:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.01619v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.01619", "catalog_id": "arxiv:2509.01619", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2025-09-01", "identifiers": { "arxiv": "2509.01619" }, "record_updated_at": "2025-09-01" }, "genre": "preprint", "id": "arxiv:2509.01619", "issued": { "date-parts": [ [ 2025, 9, 1 ] ] }, "note": "First published: 2025-09-01; Recorded arXiv version: v1; Recorded version updated: 2025-09-01T16:56:16Z.", "title": "Throttling Web Agents Using Reasoning Gates", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.03891v1", "archive": "arXiv", "archive_location": "2509.03891", "author": [ { "literal": "Gowen Loo" }, { "literal": "Chang Liu" }, { "literal": "Qinghong Yin" }, { "literal": "Xiang Chen" }, { "literal": "Jiawei Chen" }, { "literal": "Jingyuan Zhang" }, { "literal": "Yu Tian" } ], "citation-key": "arxiv_2509_03891", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a8b2db00f62afe350e55e2266b3627de9b0cf089178b0cf80eb13b06275eee15", "submitted_at": "2025-09-04T05:22:42Z", "updated_at": "2025-09-04T05:22:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.03891v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.03891", "catalog_id": "arxiv:2509.03891", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-04", "identifiers": { "arxiv": "2509.03891" }, "record_updated_at": "2025-09-04" }, "genre": "preprint", "id": "arxiv:2509.03891", "issued": { "date-parts": [ [ 2025, 9, 4 ] ] }, "note": "First published: 2025-09-04; Recorded arXiv version: v1; Recorded version updated: 2025-09-04T05:22:42Z.", "title": "MobileRAG: Enhancing Mobile Agent with Retrieval-Augmented Generation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.04908v1", "archive": "arXiv", "archive_location": "2509.04908", "author": [ { "literal": "Hongyi Jing" }, { "literal": "Jiafu Chen" }, { "literal": "Chen Rao" }, { "literal": "Ziqiang Dang" }, { "literal": "Jiajie Teng" }, { "literal": "Tianyi Chu" }, { "literal": "Juncheng Mo" }, { "literal": "Shuo Fang" }, { "literal": "Huaizhong Lin" }, { "literal": "Rui Lv" }, { "literal": "Chenguang Ma" }, { "literal": "Lei Zhao" } ], "citation-key": "arxiv_2509_04908", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f74081d31c98d18e10a5febc5de436a3788fc4156215e4be8fc54b1c53260f58", "submitted_at": "2025-09-05T08:24:12Z", "updated_at": "2025-09-05T08:24:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.04908v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.04908", "catalog_id": "arxiv:2509.04908", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-05", "identifiers": { "arxiv": "2509.04908" }, "record_updated_at": "2025-09-05" }, "genre": "preprint", "id": "arxiv:2509.04908", "issued": { "date-parts": [ [ 2025, 9, 5 ] ] }, "note": "First published: 2025-09-05; Recorded arXiv version: v1; Recorded version updated: 2025-09-05T08:24:12Z.", "title": "SparkUI-Parser: Enhancing GUI Perception with Robust Grounding and Parsing", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.06477v2", "archive": "arXiv", "archive_location": "2509.06477", "author": [ { "literal": "Pengxiang Zhao" }, { "literal": "Guangyi Liu" }, { "literal": "YaoZhen Liang" }, { "literal": "Weiqing He" }, { "literal": "Zhengxi Lu" }, { "literal": "WenHao Wang" }, { "literal": "Yuehao Huang" }, { "literal": "Yuxiang Chai" }, { "literal": "Zhaolu Kang" }, { "literal": "Yaxuan Guo" }, { "literal": "Hao Wang" }, { "literal": "Kexin Zhang" }, { "literal": "Liang Liu" }, { "literal": "Yong Liu" } ], "citation-key": "arxiv_2509_06477", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9b1e0e807f5f396bfd9108ab6cfacd48ed4780abfb792589da0b41d7d246c717", "submitted_at": "2025-09-08T09:43:48Z", "updated_at": "2026-04-15T11:26:55Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2509.06477v2" }, "canonical_source_url": "https://arxiv.org/abs/2509.06477", "catalog_id": "arxiv:2509.06477", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-08", "identifiers": { "arxiv": "2509.06477" }, "record_updated_at": "2026-04-15" }, "genre": "preprint", "id": "arxiv:2509.06477", "issued": { "date-parts": [ [ 2025, 9, 8 ] ] }, "note": "First published: 2025-09-08; Recorded arXiv version: v2; Recorded version updated: 2026-04-15T11:26:55Z.", "title": "MAS-Bench: A Unified Benchmark for Shortcut-Augmented Hybrid Mobile GUI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2509.07764v1", "archive": "arXiv", "archive_location": "2509.07764", "author": [ { "literal": "Haitao Hu" }, { "literal": "Peng Chen" }, { "literal": "Yanpeng Zhao" }, { "literal": "Yuqi Chen" } ], "citation-key": "arxiv_2509_07764", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ae582aceb31075c460a9e4a6dce9978574c3e0ada731b26c13738a4a4fafcc32", "submitted_at": "2025-09-09T13:59:00Z", "updated_at": "2025-09-09T13:59:00Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.07764v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.07764", "catalog_id": "arxiv:2509.07764", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-09", "identifiers": { "arxiv": "2509.07764" }, "record_updated_at": "2025-09-09" }, "genre": "preprint", "id": "arxiv:2509.07764", "issued": { "date-parts": [ [ 2025, 9, 9 ] ] }, "note": "First published: 2025-09-09; Recorded arXiv version: v1; Recorded version updated: 2025-09-09T13:59:00Z.", "title": "AgentSentinel: An End-to-End and Real-Time Security Defense Framework for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.10723v1", "archive": "arXiv", "archive_location": "2509.10723", "author": [ { "literal": "Jingyu Tang" }, { "literal": "Chaoran Chen" }, { "literal": "Jiawen Li" }, { "literal": "Zhiping Zhang" }, { "literal": "Bingcan Guo" }, { "literal": "Ibrahim Khalilov" }, { "literal": "Simret Araya Gebreegziabher" }, { "literal": "Bingsheng Yao" }, { "literal": "Dakuo Wang" }, { "literal": "Yanfang Ye" }, { "literal": "Tianshi Li" }, { "literal": "Ziang Xiao" }, { "literal": "Yaxing Yao" }, { "literal": "Toby Jia-Jun Li" } ], "citation-key": "arxiv_2509_10723", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "247ed053a4c8f62106d9f8e9af0801d94265ed712f4ae9f095fd8aaccbe33c59", "submitted_at": "2025-09-12T22:26:31Z", "updated_at": "2025-09-12T22:26:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.10723v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.10723", "catalog_id": "arxiv:2509.10723", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-12", "identifiers": { "arxiv": "2509.10723" }, "record_updated_at": "2025-09-12" }, "genre": "preprint", "id": "arxiv:2509.10723", "issued": { "date-parts": [ [ 2025, 9, 12 ] ] }, "note": "First published: 2025-09-12; Recorded arXiv version: v1; Recorded version updated: 2025-09-12T22:26:31Z.", "title": "Dark Patterns Meet GUI Agents: LLM Agent Susceptibility to Manipulative Interfaces and the Role of Human Oversight", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.11250v3", "archive": "arXiv", "archive_location": "2509.11250", "author": [ { "literal": "Yitong Zhang" }, { "literal": "Ximo Li" }, { "literal": "Liyi Cai" }, { "literal": "Jia Li" } ], "citation-key": "arxiv_2509_11250", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9ab96edd8e52b3a490455b46878c92c23d15c0f9e105e71dbd6848e058da42ec", "submitted_at": "2025-09-14T12:47:54Z", "updated_at": "2026-08-24T12:04:58Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2509.11250v3" }, "canonical_source_url": "https://arxiv.org/abs/2509.11250", "catalog_id": "arxiv:2509.11250", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-14", "identifiers": { "arxiv": "2509.11250" }, "record_updated_at": "2026-08-24" }, "genre": "preprint", "id": "arxiv:2509.11250", "issued": { "date-parts": [ [ 2025, 9, 14 ] ] }, "note": "First published: 2025-09-14; Recorded arXiv version: v3; Recorded version updated: 2026-08-24T12:04:58Z.", "title": "Environmental Injection Attacks against GUI Agents in Realistic Dynamic Environments", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2509.11543v2", "archive": "arXiv", "archive_location": "2509.11543", "author": [ { "literal": "Zhengxi Lu" }, { "literal": "Jiabo Ye" }, { "literal": "Fei Tang" }, { "literal": "Yongliang Shen" }, { "literal": "Haiyang Xu" }, { "literal": "Ziwei Zheng" }, { "literal": "Weiming Lu" }, { "literal": "Ming Yan" }, { "literal": "Fei Huang" }, { "literal": "Jun Xiao" }, { "literal": "Yueting Zhuang" } ], "citation-key": "arxiv_2509_11543", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1e31638e531aff49fe59bef6e80b35069b1f851ea0cbf1901130557148df4783", "submitted_at": "2025-09-15T03:24:08Z", "updated_at": "2025-09-24T15:05:34Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2509.11543v2" }, "canonical_source_url": "https://arxiv.org/abs/2509.11543", "catalog_id": "arxiv:2509.11543", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-15", "identifiers": { "arxiv": "2509.11543" }, "record_updated_at": "2025-09-24" }, "genre": "preprint", "id": "arxiv:2509.11543", "issued": { "date-parts": [ [ 2025, 9, 15 ] ] }, "note": "First published: 2025-09-15; Recorded arXiv version: v2; Recorded version updated: 2025-09-24T15:05:34Z.", "title": "UI-S1: Advancing GUI Automation via Semi-online Reinforcement Learning", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2509.13615v5", "archive": "arXiv", "archive_location": "2509.13615", "author": [ { "literal": "Zongru Wu" }, { "literal": "Rui Mao" }, { "literal": "Zhiyuan Tian" }, { "literal": "Pengzhou Cheng" }, { "literal": "Tianjie Ju" }, { "literal": "Zheng Wu" }, { "literal": "Lingzhong Dong" }, { "literal": "Haiyue Sheng" }, { "literal": "Zhuosheng Zhang" }, { "literal": "Gongshen Liu" } ], "citation-key": "arxiv_2509_13615", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "be8620b021bd9218c1143282f1c7a7ea06be2abe6fc0928a310cfc8ac3172a18", "submitted_at": "2025-09-17T01:14:14Z", "updated_at": "2026-03-18T08:12:54Z", "version": 5, "versioned_url": "https://arxiv.org/abs/2509.13615v5" }, "canonical_source_url": "https://arxiv.org/abs/2509.13615", "catalog_id": "arxiv:2509.13615", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-17", "identifiers": { "arxiv": "2509.13615" }, "record_updated_at": "2026-03-18" }, "genre": "preprint", "id": "arxiv:2509.13615", "issued": { "date-parts": [ [ 2025, 9, 17 ] ] }, "note": "First published: 2025-09-17; Recorded arXiv version: v5; Recorded version updated: 2026-03-18T08:12:54Z.", "title": "See, Think, Act: Teaching Multimodal Agents to Effectively Interact with GUI by Identifying Toggles", "type": "article", "version": "5" }, { "URL": "https://arxiv.org/abs/2509.14382v1", "archive": "arXiv", "archive_location": "2509.14382", "author": [ { "literal": "Daniel Röder" }, { "literal": "Akhil Juneja" }, { "literal": "Roland Roller" }, { "literal": "Sven Schmeier" } ], "citation-key": "arxiv_2509_14382", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ac93e8cfca06ff86fad06408051abaa6b6510a44d7ad9f2d34c478b11391a32f", "submitted_at": "2025-09-17T19:34:49Z", "updated_at": "2025-09-17T19:34:49Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.14382v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.14382", "catalog_id": "arxiv:2509.14382", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-17", "identifiers": { "arxiv": "2509.14382" }, "record_updated_at": "2025-09-17" }, "genre": "preprint", "id": "arxiv:2509.14382", "issued": { "date-parts": [ [ 2025, 9, 17 ] ] }, "note": "First published: 2025-09-17; Recorded arXiv version: v1; Recorded version updated: 2025-09-17T19:34:49Z.", "title": "Detecting Pipeline Failures through Fine-Grained Analysis of Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.21782v2", "archive": "arXiv", "archive_location": "2509.21782", "author": [ { "literal": "Junliang Liu" }, { "literal": "Jingyu Xiao" }, { "literal": "Wenxin Tang" }, { "literal": "Zhixian Wang" }, { "literal": "Zipeng Xie" }, { "literal": "Wenxuan Wang" }, { "literal": "Minrui Zhang" }, { "literal": "Shuanghe Yu" } ], "citation-key": "arxiv_2509_21782", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8fed6f081e0330c54a304c79e8a4183865aae48f0e944545029d3de9852f8028", "submitted_at": "2025-09-26T02:38:14Z", "updated_at": "2026-03-04T16:31:39Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2509.21782v2" }, "canonical_source_url": "https://arxiv.org/abs/2509.21782", "catalog_id": "arxiv:2509.21782", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-26", "identifiers": { "arxiv": "2509.21782" }, "record_updated_at": "2026-03-04" }, "genre": "preprint", "id": "arxiv:2509.21782", "issued": { "date-parts": [ [ 2025, 9, 26 ] ] }, "note": "First published: 2025-09-26; Recorded arXiv version: v2; Recorded version updated: 2026-03-04T16:31:39Z.", "title": "Benchmarking MLLM-based Web Understanding: Reasoning, Robustness and Safety", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2509.21823v2", "archive": "arXiv", "archive_location": "2509.21823", "author": [ { "literal": "Gaole Dai" }, { "literal": "Shiqi Jiang" }, { "literal": "Ting Cao" }, { "literal": "Yuqing Yang" }, { "literal": "Yuanchun Li" }, { "literal": "Rui Tan" }, { "literal": "Mo Li" }, { "literal": "Lili Qiu" } ], "citation-key": "arxiv_2509_21823", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c96f310fb0f28888c59dce71d08697b603e7a7562fdf99b51e3504b93e9e81b3", "submitted_at": "2025-09-26T03:29:36Z", "updated_at": "2026-04-15T09:19:46Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2509.21823v2" }, "canonical_source_url": "https://arxiv.org/abs/2509.21823", "catalog_id": "arxiv:2509.21823", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-26", "identifiers": { "arxiv": "2509.21823" }, "record_updated_at": "2026-04-15" }, "genre": "preprint", "id": "arxiv:2509.21823", "issued": { "date-parts": [ [ 2025, 9, 26 ] ] }, "note": "First published: 2025-09-26; Recorded arXiv version: v2; Recorded version updated: 2026-04-15T09:19:46Z.", "title": "ProRe: A Proactive Reward System for GUI Agents via Reasoner-Actor Collaboration", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2509.21982v2", "archive": "arXiv", "archive_location": "2509.21982", "author": [ { "literal": "Renqi Chen" }, { "literal": "Zeyin Tao" }, { "literal": "Jianming Guo" }, { "literal": "Jingzhe Zhu" }, { "literal": "Yiheng Peng" }, { "literal": "Qingqing Sun" }, { "literal": "Tianyi Zhang" }, { "literal": "Shuai Chen" } ], "citation-key": "arxiv_2509_21982", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d23b9f33282f77b9d4965078ca993bcf21163dc1d3210aaa4e554d4fc12b37c5", "submitted_at": "2025-09-26T07:05:01Z", "updated_at": "2026-04-13T03:14:30Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2509.21982v2" }, "canonical_source_url": "https://arxiv.org/abs/2509.21982", "catalog_id": "arxiv:2509.21982", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-26", "identifiers": { "arxiv": "2509.21982" }, "record_updated_at": "2026-04-13" }, "genre": "preprint", "id": "arxiv:2509.21982", "issued": { "date-parts": [ [ 2025, 9, 26 ] ] }, "note": "First published: 2025-09-26; Recorded arXiv version: v2; Recorded version updated: 2026-04-13T03:14:30Z.", "title": "RISK: A Framework for GUI Agents in E-commerce Risk Management", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2509.23263v3", "archive": "arXiv", "archive_location": "2509.23263", "author": [ { "literal": "Tao Xiong" }, { "literal": "Xavier Hu" }, { "literal": "Yurun Chen" }, { "literal": "Yuhang Liu" }, { "literal": "Changqiao Wu" }, { "literal": "Pengzhi Gao" }, { "literal": "Wei Liu" }, { "literal": "Jian Luan" }, { "literal": "Shengyu Zhang" } ], "citation-key": "arxiv_2509_23263", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5b5375e77f087367383bb5de96cec46d8de6550d9a44bff6965b9201b6718a05", "submitted_at": "2025-09-27T11:42:36Z", "updated_at": "2026-08-30T18:12:14Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2509.23263v3" }, "canonical_source_url": "https://arxiv.org/abs/2509.23263", "catalog_id": "arxiv:2509.23263", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-27", "identifiers": { "arxiv": "2509.23263" }, "record_updated_at": "2026-08-30" }, "genre": "preprint", "id": "arxiv:2509.23263", "issued": { "date-parts": [ [ 2025, 9, 27 ] ] }, "note": "First published: 2025-09-27; Recorded arXiv version: v3; Recorded version updated: 2026-08-30T18:12:14Z.", "title": "GUI-PRA: Process Reward Agent for GUI Tasks", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2509.24133v1", "archive": "arXiv", "archive_location": "2509.24133", "author": [ { "literal": "Zhecheng Li" }, { "literal": "Guoxian Song" }, { "literal": "Yiwei Wang" }, { "literal": "Zhen Xiong" }, { "literal": "Junsong Yuan" }, { "literal": "Yujun Cai" } ], "citation-key": "arxiv_2509_24133", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9b42e784de8dd007c60d234b7331bd98b5a04c58c13696bf436c81c7eccee37f", "submitted_at": "2025-09-29T00:06:31Z", "updated_at": "2025-09-29T00:06:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.24133v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.24133", "catalog_id": "arxiv:2509.24133", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-29", "identifiers": { "arxiv": "2509.24133" }, "record_updated_at": "2025-09-29" }, "genre": "preprint", "id": "arxiv:2509.24133", "issued": { "date-parts": [ [ 2025, 9, 29 ] ] }, "note": "First published: 2025-09-29; Recorded arXiv version: v1; Recorded version updated: 2025-09-29T00:06:31Z.", "title": "Generalist Scanner Meets Specialist Locator: A Synergistic Coarse-to-Fine Framework for Robust GUI Grounding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2509.26506v1", "archive": "arXiv", "archive_location": "2509.26506", "author": [ { "literal": "Yutong Dai" }, { "literal": "Krithika Ramakrishnan" }, { "literal": "Jing Gu" }, { "literal": "Matthew Fernandez" }, { "literal": "Yanqi Luo" }, { "literal": "Viraj Prabhu" }, { "literal": "Zhenyu Hu" }, { "literal": "Silvio Savarese" }, { "literal": "Caiming Xiong" }, { "literal": "Zeyuan Chen" }, { "literal": "Ran Xu" } ], "citation-key": "arxiv_2509_26506", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0e22d26dfbda8530f32d4686f26d91ea1a8e8c5470e857f2e13803258a2f7aaf", "submitted_at": "2025-09-30T16:48:49Z", "updated_at": "2025-09-30T16:48:49Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2509.26506v1" }, "canonical_source_url": "https://arxiv.org/abs/2509.26506", "catalog_id": "arxiv:2509.26506", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-30", "identifiers": { "arxiv": "2509.26506" }, "record_updated_at": "2025-09-30" }, "genre": "preprint", "id": "arxiv:2509.26506", "issued": { "date-parts": [ [ 2025, 9, 30 ] ] }, "note": "First published: 2025-09-30; Recorded arXiv version: v1; Recorded version updated: 2025-09-30T16:48:49Z.", "title": "SCUBA: Salesforce Computer Use Benchmark", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.01354v2", "archive": "arXiv", "archive_location": "2510.01354", "author": [ { "literal": "Yinuo Liu" }, { "literal": "Xilong Wang" }, { "literal": "Ruohan Xu" }, { "literal": "Yuqi Jia" }, { "literal": "Neil Zhenqiang Gong" } ], "citation-key": "arxiv_2510_01354", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e997e87268cfe66869bf19fdc8445e5487c980736a63cbe76f779bd1f142c396", "submitted_at": "2025-10-01T18:34:06Z", "updated_at": "2026-09-23T16:44:27Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.01354v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.01354", "catalog_id": "arxiv:2510.01354", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-01", "identifiers": { "arxiv": "2510.01354" }, "record_updated_at": "2026-09-23" }, "genre": "preprint", "id": "arxiv:2510.01354", "issued": { "date-parts": [ [ 2025, 10, 1 ] ] }, "note": "First published: 2025-10-01; Recorded arXiv version: v2; Recorded version updated: 2026-09-23T16:44:27Z.", "title": "WAInjectBench: Benchmarking Prompt Injection Detections for Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.01670v1", "archive": "arXiv", "archive_location": "2510.01670", "author": [ { "literal": "Erfan Shayegani" }, { "literal": "Keegan Hines" }, { "literal": "Yue Dong" }, { "literal": "Nael Abu-Ghazaleh" }, { "literal": "Roman Lutz" }, { "literal": "Spencer Whitehead" }, { "literal": "Vidhisha Balachandran" }, { "literal": "Besmira Nushi" }, { "literal": "Vibhav Vineet" } ], "citation-key": "arxiv_2510_01670", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a6426fa5eeb3dfafef30e47dd46a6a43dff3e24bf0ff485b7d10ab9efe133f94", "submitted_at": "2025-10-02T04:52:15Z", "updated_at": "2025-10-02T04:52:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.01670v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.01670", "catalog_id": "arxiv:2510.01670", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-02", "identifiers": { "arxiv": "2510.01670" }, "record_updated_at": "2025-10-02" }, "genre": "preprint", "id": "arxiv:2510.01670", "issued": { "date-parts": [ [ 2025, 10, 2 ] ] }, "note": "First published: 2025-10-02; Recorded arXiv version: v1; Recorded version updated: 2025-10-02T04:52:15Z.", "title": "Just Do It!? Computer-Use Agents Exhibit Blind Goal-Directedness", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.02250v2", "archive": "arXiv", "archive_location": "2510.02250", "author": [ { "literal": "Gonzalo Gonzalez-Pumariega" }, { "literal": "Vincent Tu" }, { "literal": "Chih-Lun Lee" }, { "literal": "Jiachen Yang" }, { "literal": "Ang Li" }, { "literal": "Xin Eric Wang" } ], "citation-key": "arxiv_2510_02250", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "29c00ba4713da26f912fec8046d498b4fd10cf71a3109f66e92fa23f7cebcf44", "submitted_at": "2025-10-02T17:37:08Z", "updated_at": "2026-02-03T23:48:45Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.02250v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.02250", "catalog_id": "arxiv:2510.02250", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-02", "identifiers": { "arxiv": "2510.02250" }, "record_updated_at": "2026-02-03" }, "genre": "preprint", "id": "arxiv:2510.02250", "issued": { "date-parts": [ [ 2025, 10, 2 ] ] }, "note": "First published: 2025-10-02; Recorded arXiv version: v2; Recorded version updated: 2026-02-03T23:48:45Z.", "title": "Scaling Agents for Computer Use", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.02418v2", "archive": "arXiv", "archive_location": "2510.02418", "author": [ { "literal": "Sagnik Anupam" }, { "literal": "Davis Brown" }, { "literal": "Shuo Li" }, { "literal": "Eric Wong" }, { "literal": "Hamed Hassani" }, { "literal": "Osbert Bastani" } ], "citation-key": "arxiv_2510_02418", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7906e06c2efb8778a43ec774821f90947e70cec0b78d6d43ac8901f7f0562332", "submitted_at": "2025-10-02T15:22:21Z", "updated_at": "2025-10-07T15:12:39Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.02418v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.02418", "catalog_id": "arxiv:2510.02418", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-02", "identifiers": { "arxiv": "2510.02418" }, "record_updated_at": "2025-10-07" }, "genre": "preprint", "id": "arxiv:2510.02418", "issued": { "date-parts": [ [ 2025, 10, 2 ] ] }, "note": "First published: 2025-10-02; Recorded arXiv version: v2; Recorded version updated: 2025-10-07T15:12:39Z.", "title": "BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.03285v1", "archive": "arXiv", "archive_location": "2510.03285", "author": [ { "literal": "Su Kara" }, { "literal": "Fazle Faisal" }, { "literal": "Suman Nath" } ], "citation-key": "arxiv_2510_03285", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "88e25f6d876dc6ad11219e3faa0e7d95f4abb39a9aba32213896225aff64ec42", "submitted_at": "2025-09-28T20:51:05Z", "updated_at": "2025-09-28T20:51:05Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.03285v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.03285", "catalog_id": "arxiv:2510.03285", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-09-28", "identifiers": { "arxiv": "2510.03285" }, "record_updated_at": "2025-09-28" }, "genre": "preprint", "id": "arxiv:2510.03285", "issued": { "date-parts": [ [ 2025, 9, 28 ] ] }, "note": "First published: 2025-09-28; Recorded arXiv version: v1; Recorded version updated: 2025-09-28T20:51:05Z.", "title": "WAREX: Web Agent Reliability Evaluation on Existing Benchmarks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.09872v2", "archive": "arXiv", "archive_location": "2510.09872", "author": [ { "literal": "Sanjari Srivastava" }, { "literal": "Gang Li" }, { "literal": "Cheng Chang" }, { "literal": "Rishu Garg" }, { "literal": "Manpreet Kaur" }, { "literal": "Charlene Y. Lee" }, { "literal": "Yuezhang Li" }, { "literal": "Yining Mao" }, { "literal": "Ignacio Cases" }, { "literal": "Yanan Xie" }, { "literal": "Peng Qi" } ], "citation-key": "arxiv_2510_09872", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3af3fb75d012ebcf4ad082ab7b83b0553596bbbb8598751d0a70c86d929a22ee", "submitted_at": "2025-10-10T21:20:51Z", "updated_at": "2026-05-18T22:06:26Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.09872v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.09872", "catalog_id": "arxiv:2510.09872", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-10", "identifiers": { "arxiv": "2510.09872" }, "record_updated_at": "2026-05-18" }, "genre": "preprint", "id": "arxiv:2510.09872", "issued": { "date-parts": [ [ 2025, 10, 10 ] ] }, "note": "First published: 2025-10-10; Recorded arXiv version: v2; Recorded version updated: 2026-05-18T22:06:26Z.", "title": "WARC-Bench: Web Archive Based Benchmark for GUI Subtask Executions", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.10073v2", "archive": "arXiv", "archive_location": "2510.10073", "author": [ { "literal": "Zonghao Ying" }, { "literal": "Yangguang Shao" }, { "literal": "Jianle Gan" }, { "literal": "Gan Xu" }, { "literal": "Wenxin Zhang" }, { "literal": "Quanchen Zou" }, { "literal": "Junzheng Shi" }, { "literal": "Zhenfei Yin" }, { "literal": "Mingchuan Zhang" }, { "literal": "Aishan Liu" }, { "literal": "Xianglong Liu" } ], "citation-key": "arxiv_2510_10073", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "43c9f7c91539df587a4f08888820731aef5bb3fca2cfb9e742443330aaf1c35f", "submitted_at": "2025-10-11T07:18:12Z", "updated_at": "2026-04-14T02:53:37Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.10073v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.10073", "catalog_id": "arxiv:2510.10073", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-11", "identifiers": { "arxiv": "2510.10073" }, "record_updated_at": "2026-04-14" }, "genre": "preprint", "id": "arxiv:2510.10073", "issued": { "date-parts": [ [ 2025, 10, 11 ] ] }, "note": "First published: 2025-10-11; Recorded arXiv version: v2; Recorded version updated: 2026-04-14T02:53:37Z.", "title": "SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.11035v2", "archive": "arXiv", "archive_location": "2510.11035", "author": [ { "literal": "Longjie Guo" }, { "literal": "Chenjie Yuan" }, { "literal": "Mingyuan Zhong" }, { "literal": "Robert Wolfe" }, { "literal": "Ruican Zhong" }, { "literal": "Yue Xu" }, { "literal": "Bingbing Wen" }, { "literal": "Hua Shen" }, { "literal": "Lucy Lu Wang" }, { "literal": "Alexis Hiniker" } ], "citation-key": "arxiv_2510_11035", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e6152a61f81c435e76a432bc10086827c0cf0d2f44026bac03d6b80ebae8ca77", "submitted_at": "2025-10-13T06:12:03Z", "updated_at": "2026-02-23T08:31:13Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.11035v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.11035", "catalog_id": "arxiv:2510.11035", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-13", "identifiers": { "arxiv": "2510.11035" }, "record_updated_at": "2026-02-23" }, "genre": "preprint", "id": "arxiv:2510.11035", "issued": { "date-parts": [ [ 2025, 10, 13 ] ] }, "note": "First published: 2025-10-13; Recorded arXiv version: v2; Recorded version updated: 2026-02-23T08:31:13Z.", "title": "SusBench: An Online Benchmark for Evaluating Dark Pattern Susceptibility of Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.11892v2", "archive": "arXiv", "archive_location": "2510.11892", "author": [ { "literal": "Kai Mei" }, { "literal": "Jiang Guo" }, { "literal": "Shuaichen Chang" }, { "literal": "Mingwen Dong" }, { "literal": "Dongkyu Lee" }, { "literal": "Xing Niu" }, { "literal": "Jiarong Jiang" } ], "citation-key": "arxiv_2510_11892", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7c37dae372655e8ef556163744e47ed42db3533175ca7d1db7560033774ed2a0", "submitted_at": "2025-10-13T19:52:04Z", "updated_at": "2026-03-09T15:25:16Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.11892v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.11892", "catalog_id": "arxiv:2510.11892", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-13", "identifiers": { "arxiv": "2510.11892" }, "record_updated_at": "2026-03-09" }, "genre": "preprint", "id": "arxiv:2510.11892", "issued": { "date-parts": [ [ 2025, 10, 13 ] ] }, "note": "First published: 2025-10-13; Recorded arXiv version: v2; Recorded version updated: 2026-03-09T15:25:16Z.", "title": "R-WoM: Retrieval-augmented World Model For Computer-use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.12200v1", "archive": "arXiv", "archive_location": "2510.12200", "author": [ { "literal": "Xiaoxue Ren" }, { "literal": "Penghao Jiang" }, { "literal": "Kaixin Li" }, { "literal": "Zhiyong Huang" }, { "literal": "Xiaoning Du" }, { "literal": "Jiaojiao Jiang" }, { "literal": "Zhenchang Xing" }, { "literal": "Jiamou Sun" }, { "literal": "Terry Yue Zhuo" } ], "citation-key": "arxiv_2510_12200", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e92db1e568c3c512aa00e2363bbdc9869cf834d638c331b45d4753d29a282004", "submitted_at": "2025-10-14T06:52:15Z", "updated_at": "2025-10-14T06:52:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.12200v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.12200", "catalog_id": "arxiv:2510.12200", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-14", "identifiers": { "arxiv": "2510.12200" }, "record_updated_at": "2025-10-14" }, "genre": "preprint", "id": "arxiv:2510.12200", "issued": { "date-parts": [ [ 2025, 10, 14 ] ] }, "note": "First published: 2025-10-14; Recorded arXiv version: v1; Recorded version updated: 2025-10-14T06:52:15Z.", "title": "HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.14621v1", "archive": "arXiv", "archive_location": "2510.14621", "author": [ { "literal": "Yuanyi Song" }, { "literal": "Heyuan Huang" }, { "literal": "Qiqiang Lin" }, { "literal": "Yin Zhao" }, { "literal": "Xiangmou Qu" }, { "literal": "Jun Wang" }, { "literal": "Xingyu Lou" }, { "literal": "Weiwen Liu" }, { "literal": "Zhuosheng Zhang" }, { "literal": "Jun Wang" }, { "literal": "Yong Yu" }, { "literal": "Weinan Zhang" }, { "literal": "Zhaoxiang Wang" } ], "citation-key": "arxiv_2510_14621", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fc45dc62068c971cd831fc3db64bd3d8995d4e5fd60e9c83953476d4fb24f7f3", "submitted_at": "2025-10-16T12:30:05Z", "updated_at": "2025-10-16T12:30:05Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.14621v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.14621", "catalog_id": "arxiv:2510.14621", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-16", "identifiers": { "arxiv": "2510.14621" }, "record_updated_at": "2025-10-16" }, "genre": "preprint", "id": "arxiv:2510.14621", "issued": { "date-parts": [ [ 2025, 10, 16 ] ] }, "note": "First published: 2025-10-16; Recorded arXiv version: v1; Recorded version updated: 2025-10-16T12:30:05Z.", "title": "ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.18113v1", "archive": "arXiv", "archive_location": "2510.18113", "author": [ { "literal": "Devin Ersoy" }, { "literal": "Brandon Lee" }, { "literal": "Ananth Shreekumar" }, { "literal": "Arjun Arunasalam" }, { "literal": "Muhammad Ibrahim" }, { "literal": "Antonio Bianchi" }, { "literal": "Z. Berkay Celik" } ], "citation-key": "arxiv_2510_18113", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "db3c25ba55221ca3d7c45ce4243a4a37b56a698909c0ea3eaaabdcd37388a1ec", "submitted_at": "2025-10-20T21:26:26Z", "updated_at": "2025-10-20T21:26:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.18113v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.18113", "catalog_id": "arxiv:2510.18113", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-20", "identifiers": { "arxiv": "2510.18113" }, "record_updated_at": "2025-10-20" }, "genre": "preprint", "id": "arxiv:2510.18113", "issued": { "date-parts": [ [ 2025, 10, 20 ] ] }, "note": "First published: 2025-10-20; Recorded arXiv version: v1; Recorded version updated: 2025-10-20T21:26:26Z.", "title": "Investigating the Impact of Dark Patterns on LLM-Based Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.18488v1", "archive": "arXiv", "archive_location": "2510.18488", "author": [ { "literal": "Ho Fai Leung" }, { "literal": "Xiaoyan Xi" }, { "literal": "Fei Zuo" } ], "citation-key": "arxiv_2510_18488", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "33b21a63d51148f2b5c30478c302e3df9b92e03e2e106a1c68224f2c835a8608", "submitted_at": "2025-10-21T10:11:33Z", "updated_at": "2025-10-21T10:11:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.18488v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.18488", "catalog_id": "arxiv:2510.18488", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-21", "identifiers": { "arxiv": "2510.18488" }, "record_updated_at": "2025-10-21" }, "genre": "preprint", "id": "arxiv:2510.18488", "issued": { "date-parts": [ [ 2025, 10, 21 ] ] }, "note": "First published: 2025-10-21; Recorded arXiv version: v1; Recorded version updated: 2025-10-21T10:11:33Z.", "title": "AndroidControl-Curated: Revealing the True Potential of GUI Agents through Benchmark Purification", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.19205v1", "archive": "arXiv", "archive_location": "2510.19205", "author": [ { "literal": "Yaoyao Qian" }, { "literal": "Yuanli Wang" }, { "literal": "Jinda Zhang" }, { "literal": "Yun Zong" }, { "literal": "Meixu Chen" }, { "literal": "Hanhan Zhou" }, { "literal": "Jindan Huang" }, { "literal": "Yifan Zeng" }, { "literal": "Xinyu Hu" }, { "literal": "Chan Hee Song" }, { "literal": "Danqing Zhang" } ], "citation-key": "arxiv_2510_19205", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d404ff922ef38e57481c01a5e3108bd38457d88f83d7adf6d3c3c7b5ddf124e2", "submitted_at": "2025-10-22T03:29:25Z", "updated_at": "2025-10-22T03:29:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2510.19205v1" }, "canonical_source_url": "https://arxiv.org/abs/2510.19205", "catalog_id": "arxiv:2510.19205", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-22", "identifiers": { "arxiv": "2510.19205" }, "record_updated_at": "2025-10-22" }, "genre": "preprint", "id": "arxiv:2510.19205", "issued": { "date-parts": [ [ 2025, 10, 22 ] ] }, "note": "First published: 2025-10-22; Recorded arXiv version: v1; Recorded version updated: 2025-10-22T03:29:25Z.", "title": "WebGraphEval: Multi-Turn Trajectory Evaluation for Web Agents using Graph Representation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2510.24411v3", "archive": "arXiv", "archive_location": "2510.24411", "author": [ { "literal": "Qiushi Sun" }, { "literal": "Mukai Li" }, { "literal": "Zhoumianze Liu" }, { "literal": "Zhihui Xie" }, { "literal": "Fangzhi Xu" }, { "literal": "Zhangyue Yin" }, { "literal": "Kanzhi Cheng" }, { "literal": "Zehao Li" }, { "literal": "Zichen Ding" }, { "literal": "Qi Liu" }, { "literal": "Zhiyong Wu" }, { "literal": "Zhuosheng Zhang" }, { "literal": "Ben Kao" }, { "literal": "Lingpeng Kong" } ], "citation-key": "arxiv_2510_24411", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7e3fde732c9e760437f2c60d585f7ba0d2ed13bbe9e051838844a21174159099", "submitted_at": "2025-10-28T13:22:39Z", "updated_at": "2026-07-26T01:16:06Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2510.24411v3" }, "canonical_source_url": "https://arxiv.org/abs/2510.24411", "catalog_id": "arxiv:2510.24411", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-28", "identifiers": { "arxiv": "2510.24411" }, "record_updated_at": "2026-07-26" }, "genre": "preprint", "id": "arxiv:2510.24411", "issued": { "date-parts": [ [ 2025, 10, 28 ] ] }, "note": "First published: 2025-10-28; Recorded arXiv version: v3; Recorded version updated: 2026-07-26T01:16:06Z.", "title": "OS-Sentinel: Towards Safety-Enhanced Mobile GUI Agents via Hybrid Validation in Realistic Workflows", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2510.24563v2", "archive": "arXiv", "archive_location": "2510.24563", "author": [ { "literal": "Hongrui Jia" }, { "literal": "Jitong Liao" }, { "literal": "Xi Zhang" }, { "literal": "Haiyang Xu" }, { "literal": "Tianbao Xie" }, { "literal": "Chaoya Jiang" }, { "literal": "Ming Yan" }, { "literal": "Si Liu" }, { "literal": "Wei Ye" }, { "literal": "Fei Huang" } ], "citation-key": "arxiv_2510_24563", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d115356f723746ce95c3e40fed933786a0747b3b560c8f73a6b7af05ca22d24f", "submitted_at": "2025-10-28T15:56:36Z", "updated_at": "2025-11-11T04:10:32Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.24563v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.24563", "catalog_id": "arxiv:2510.24563", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-28", "identifiers": { "arxiv": "2510.24563" }, "record_updated_at": "2025-11-11" }, "genre": "preprint", "id": "arxiv:2510.24563", "issued": { "date-parts": [ [ 2025, 10, 28 ] ] }, "note": "First published: 2025-10-28; Recorded arXiv version: v2; Recorded version updated: 2025-11-11T04:10:32Z.", "title": "OSWorld-MCP: Benchmarking MCP Tool Invocation In Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2510.27140v3", "archive": "arXiv", "archive_location": "2510.27140", "author": [ { "literal": "Chenghao Du" }, { "literal": "Quanfeng Huang" }, { "literal": "Tingxuan Tang" }, { "literal": "Zihao Wang" }, { "literal": "Adwait Nadkarni" }, { "literal": "Yue Xiao" } ], "citation-key": "arxiv_2510_27140", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "61b611712fff54af25aae660f2353fa6c3aaa764b0767831ebf2278c120d9dea", "submitted_at": "2025-10-31T03:35:59Z", "updated_at": "2026-07-07T02:18:26Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2510.27140v3" }, "canonical_source_url": "https://arxiv.org/abs/2510.27140", "catalog_id": "arxiv:2510.27140", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-31", "identifiers": { "arxiv": "2510.27140" }, "record_updated_at": "2026-07-07" }, "genre": "preprint", "id": "arxiv:2510.27140", "issued": { "date-parts": [ [ 2025, 10, 31 ] ] }, "note": "First published: 2025-10-31; Recorded arXiv version: v3; Recorded version updated: 2026-07-07T02:18:26Z.", "title": "Measuring the Security of Mobile LLM Agents under Adversarial Prompts from Untrusted Third-Party Channels", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2510.27266v2", "archive": "arXiv", "archive_location": "2510.27266", "author": [ { "literal": "Shaojie Zhang" }, { "literal": "Pei Fu" }, { "literal": "Ruoceng Zhang" }, { "literal": "Jiahui Yang" }, { "literal": "Anan Du" }, { "literal": "Xiuwen Xi" }, { "literal": "Shaokang Wang" }, { "literal": "Ying Huang" }, { "literal": "Bin Qin" }, { "literal": "Zhenbo Luo" }, { "literal": "Jian Luan" } ], "citation-key": "arxiv_2510_27266", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6869e86d7cd4242ba8dcf942bb1cef3c157dff4f0ef37f7138ff5025d082c7b8", "submitted_at": "2025-10-31T08:07:02Z", "updated_at": "2026-05-27T02:32:58Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2510.27266v2" }, "canonical_source_url": "https://arxiv.org/abs/2510.27266", "catalog_id": "arxiv:2510.27266", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-10-31", "identifiers": { "arxiv": "2510.27266" }, "record_updated_at": "2026-05-27" }, "genre": "preprint", "id": "arxiv:2510.27266", "issued": { "date-parts": [ [ 2025, 10, 31 ] ] }, "note": "First published: 2025-10-31; Recorded arXiv version: v2; Recorded version updated: 2026-05-27T02:32:58Z.", "title": "Enhancing Trustworthy GUI Grounding via Self-Critiqued Reinforcement Learning", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2511.04307v2", "archive": "arXiv", "archive_location": "2511.04307", "author": [ { "literal": "Jian Mu" }, { "literal": "Chaoyun Zhang" }, { "literal": "Chiming Ni" }, { "literal": "Lu Wang" }, { "literal": "Bo Qiao" }, { "literal": "Kartik Mathur" }, { "literal": "Qianhui Wu" }, { "literal": "Yuhang Xie" }, { "literal": "Xiaojun Ma" }, { "literal": "Mengyu Zhou" }, { "literal": "Si Qin" }, { "literal": "Liqun Li" }, { "literal": "Yu Kang" }, { "literal": "Minghua Ma" }, { "literal": "Qingwei Lin" }, { "literal": "Saravan Rajmohan" }, { "literal": "Dongmei Zhang" } ], "citation-key": "arxiv_2511_04307", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6b063a4d2b3a552aa05a153745978473a7d165f2c6b4b97e4c91eff11b7b0f80", "submitted_at": "2025-11-06T12:19:02Z", "updated_at": "2025-11-10T12:27:15Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2511.04307v2" }, "canonical_source_url": "https://arxiv.org/abs/2511.04307", "catalog_id": "arxiv:2511.04307", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-06", "identifiers": { "arxiv": "2511.04307" }, "record_updated_at": "2025-11-10" }, "genre": "preprint", "id": "arxiv:2511.04307", "issued": { "date-parts": [ [ 2025, 11, 6 ] ] }, "note": "First published: 2025-11-06; Recorded arXiv version: v2; Recorded version updated: 2025-11-10T12:27:15Z.", "title": "GUI-360$^\\circ$: A Comprehensive Dataset and Benchmark for Computer-Using Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2511.06417v1", "archive": "arXiv", "archive_location": "2511.06417", "author": [ { "literal": "Xiangwu Guo" }, { "literal": "Difei Gao" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2511_06417", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8a5f43873a47e5b47250cedd3d860abe2dc5ec9b85fe657444bf2b7378f3e8f7", "submitted_at": "2025-11-09T15:13:45Z", "updated_at": "2025-11-09T15:13:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.06417v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.06417", "catalog_id": "arxiv:2511.06417", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-09", "identifiers": { "arxiv": "2511.06417" }, "record_updated_at": "2025-11-09" }, "genre": "preprint", "id": "arxiv:2511.06417", "issued": { "date-parts": [ [ 2025, 11, 9 ] ] }, "note": "First published: 2025-11-09; Recorded arXiv version: v1; Recorded version updated: 2025-11-09T15:13:45Z.", "title": "AUTO-Explorer: Automated Data Collection for GUI Agent", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2511.07332v2", "archive": "arXiv", "archive_location": "2511.07332", "author": [ { "literal": "Aarash Feizi" }, { "literal": "Shravan Nayak" }, { "literal": "Xiangru Jian" }, { "literal": "Kevin Qinghong Lin" }, { "literal": "Kaixin Li" }, { "literal": "Rabiul Awal" }, { "literal": "Xing Han Lù" }, { "literal": "Johan Obando-Ceron" }, { "literal": "Juan A. Rodriguez" }, { "literal": "Nicolas Chapados" }, { "literal": "David Vazquez" }, { "literal": "Adriana Romero-Soriano" }, { "literal": "Reihaneh Rabbany" }, { "literal": "Perouz Taslakian" }, { "literal": "Christopher Pal" }, { "literal": "Spandana Gella" }, { "literal": "Sai Rajeswar" } ], "citation-key": "arxiv_2511_07332", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0f2b80f294ddf5e3331dcb47a836864032f88f2fdbe35588bfef000cd9ceddc8", "submitted_at": "2025-11-10T17:35:21Z", "updated_at": "2026-06-09T21:30:32Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2511.07332v2" }, "canonical_source_url": "https://arxiv.org/abs/2511.07332", "catalog_id": "arxiv:2511.07332", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-10", "identifiers": { "arxiv": "2511.07332" }, "record_updated_at": "2026-06-09" }, "genre": "preprint", "id": "arxiv:2511.07332", "issued": { "date-parts": [ [ 2025, 11, 10 ] ] }, "note": "First published: 2025-11-10; Recorded arXiv version: v2; Recorded version updated: 2026-06-09T21:30:32Z.", "title": "Grounding Computer Use Agents on Human Demonstrations", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2511.09157v1", "archive": "arXiv", "archive_location": "2511.09157", "author": [ { "literal": "Leyang Yang" }, { "literal": "Ziwei Wang" }, { "literal": "Xiaoxuan Tang" }, { "literal": "Sheng Zhou" }, { "literal": "Dajun Chen" }, { "literal": "Wei Jiang" }, { "literal": "Yong Li" } ], "citation-key": "arxiv_2511_09157", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "dc7f5f0fe1ca373272cda4bbeca3a465e1a7eaec829e8a59bbecbe6049dc7b4a", "submitted_at": "2025-11-12T09:49:31Z", "updated_at": "2025-11-12T09:49:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.09157v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.09157", "catalog_id": "arxiv:2511.09157", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-12", "identifiers": { "arxiv": "2511.09157" }, "record_updated_at": "2025-11-12" }, "genre": "preprint", "id": "arxiv:2511.09157", "issued": { "date-parts": [ [ 2025, 11, 12 ] ] }, "note": "First published: 2025-11-12; Recorded arXiv version: v1; Recorded version updated: 2025-11-12T09:49:31Z.", "title": "ProBench: Benchmarking GUI Agents with Accurate Process Information", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2511.15567v1", "archive": "arXiv", "archive_location": "2511.15567", "author": [ { "literal": "Kevin Qinghong Lin" }, { "literal": "Siyuan Hu" }, { "literal": "Linjie Li" }, { "literal": "Zhengyuan Yang" }, { "literal": "Lijuan Wang" }, { "literal": "Philip Torr" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2511_15567", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7d63831bc0ff00979615415ce94abc140b646a4375cb33b950c4a7ccf00bf166", "submitted_at": "2025-11-19T16:00:02Z", "updated_at": "2025-11-19T16:00:02Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.15567v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.15567", "catalog_id": "arxiv:2511.15567", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-19", "identifiers": { "arxiv": "2511.15567" }, "record_updated_at": "2025-11-19" }, "genre": "preprint", "id": "arxiv:2511.15567", "issued": { "date-parts": [ [ 2025, 11, 19 ] ] }, "note": "First published: 2025-11-19; Recorded arXiv version: v1; Recorded version updated: 2025-11-19T16:00:02Z.", "title": "Computer-Use Agents as Judges for Generative User Interface", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2511.16590v1", "archive": "arXiv", "archive_location": "2511.16590", "author": [ { "literal": "Sen Chen" }, { "literal": "Tong Zhao" }, { "literal": "Yi Bin" }, { "literal": "Fei Ma" }, { "literal": "Wenqi Shao" }, { "literal": "Zheng Wang" } ], "citation-key": "arxiv_2511_16590", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1679f7b56c7b60f92195e488c887114bcdcce7975a80fafbf2572c2b1f01ce0b", "submitted_at": "2025-11-20T17:43:46Z", "updated_at": "2025-11-20T17:43:46Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.16590v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.16590", "catalog_id": "arxiv:2511.16590", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-20", "identifiers": { "arxiv": "2511.16590" }, "record_updated_at": "2025-11-20" }, "genre": "preprint", "id": "arxiv:2511.16590", "issued": { "date-parts": [ [ 2025, 11, 20 ] ] }, "note": "First published: 2025-11-20; Recorded arXiv version: v1; Recorded version updated: 2025-11-20T17:43:46Z.", "title": "D-GARA: A Dynamic Benchmarking Framework for GUI Agent Robustness in Real-World Anomalies", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2511.17131v1", "archive": "arXiv", "archive_location": "2511.17131", "author": [ { "literal": "Horia Cristescu" }, { "literal": "Charles Park" }, { "literal": "Trong Canh Nguyen" }, { "literal": "Sergiu Talmacel" }, { "literal": "Alexandru-Gabriel Ilie" }, { "literal": "Stefan Adam" } ], "citation-key": "arxiv_2511_17131", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4c746b329229d90dbe3d04a138f54c8b205b16ec6173cf681d47a319c8f24ddf", "submitted_at": "2025-11-21T10:47:22Z", "updated_at": "2025-11-21T10:47:22Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.17131v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.17131", "catalog_id": "arxiv:2511.17131", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-21", "identifiers": { "arxiv": "2511.17131" }, "record_updated_at": "2025-11-21" }, "genre": "preprint", "id": "arxiv:2511.17131", "issued": { "date-parts": [ [ 2025, 11, 21 ] ] }, "note": "First published: 2025-11-21; Recorded arXiv version: v1; Recorded version updated: 2025-11-21T10:47:22Z.", "title": "UI-CUBE: Enterprise-Grade Computer Use Agent Benchmarking Beyond Task Accuracy to Operational Reliability", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2511.20067v1", "archive": "arXiv", "archive_location": "2511.20067", "author": [ { "literal": "Marta Sumyk" }, { "literal": "Oleksandr Kosovan" } ], "citation-key": "arxiv_2511_20067", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ff9343bb706687caf2ea813220ee7221819cd5892ba950f3168b91aebe0a74e0", "submitted_at": "2025-11-25T08:40:33Z", "updated_at": "2025-11-25T08:40:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.20067v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.20067", "catalog_id": "arxiv:2511.20067", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-25", "identifiers": { "arxiv": "2511.20067" }, "record_updated_at": "2025-11-25" }, "genre": "preprint", "id": "arxiv:2511.20067", "issued": { "date-parts": [ [ 2025, 11, 25 ] ] }, "note": "First published: 2025-11-25; Recorded arXiv version: v1; Recorded version updated: 2025-11-25T08:40:33Z.", "title": "\"Are We Done Yet?\": A Vision-Based Judge for Autonomous Task Completion of Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2511.20766v1", "archive": "arXiv", "archive_location": "2511.20766", "author": [ { "literal": "Karen Ullrich" }, { "literal": "Jingtong Su" }, { "literal": "Claudia Shi" }, { "literal": "Arjun Subramonian" }, { "literal": "Amir Bar" }, { "literal": "Ivan Evtimov" }, { "literal": "Nikolaos Tsilivis" }, { "literal": "Randall Balestriero" }, { "literal": "Julia Kempe" }, { "literal": "Mark Ibrahim" } ], "citation-key": "arxiv_2511_20766", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2c288e7c42b5eedeea3f27a5622e87981f101f472db05ead6d6bede1371500f0", "submitted_at": "2025-11-25T19:00:22Z", "updated_at": "2025-11-25T19:00:22Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2511.20766v1" }, "canonical_source_url": "https://arxiv.org/abs/2511.20766", "catalog_id": "arxiv:2511.20766", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-25", "identifiers": { "arxiv": "2511.20766" }, "record_updated_at": "2025-11-25" }, "genre": "preprint", "id": "arxiv:2511.20766", "issued": { "date-parts": [ [ 2025, 11, 25 ] ] }, "note": "First published: 2025-11-25; Recorded arXiv version: v1; Recorded version updated: 2025-11-25T19:00:22Z.", "title": "OpenApps: Simulating Environment Variations to Measure UI-Agent Reliability", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.00756v2", "archive": "arXiv", "archive_location": "2512.00756", "author": [ { "literal": "Ruihan Chen" }, { "literal": "Qiming Li" }, { "literal": "Xiaocheng Feng" }, { "literal": "Weihong Zhong" }, { "literal": "Xiaoliang Yang" }, { "literal": "Yuxuan Gu" }, { "literal": "Zekun Zhou" }, { "literal": "Yunfei Lu" }, { "literal": "Haoyu Ren" }, { "literal": "Kun Chen" }, { "literal": "Dandan Tu" }, { "literal": "Bing Qin" } ], "citation-key": "arxiv_2512_00756", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bf16987178fc24fb483bfa82efecb1924c7528fd16ac94f1144fcc345748850f", "submitted_at": "2025-11-30T06:47:33Z", "updated_at": "2026-04-28T03:32:32Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2512.00756v2" }, "canonical_source_url": "https://arxiv.org/abs/2512.00756", "catalog_id": "arxiv:2512.00756", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-30", "identifiers": { "arxiv": "2512.00756" }, "record_updated_at": "2026-04-28" }, "genre": "preprint", "id": "arxiv:2512.00756", "issued": { "date-parts": [ [ 2025, 11, 30 ] ] }, "note": "First published: 2025-11-30; Recorded arXiv version: v2; Recorded version updated: 2026-04-28T03:32:32Z.", "title": "MPR-GUI: Benchmarking and Enhancing Multilingual Perception and Reasoning in GUI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2512.01174v1", "archive": "arXiv", "archive_location": "2512.01174", "author": [ { "literal": "Hyunjun Kim" }, { "literal": "Sooyoung Ryu" } ], "citation-key": "arxiv_2512_01174", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7e445d0ca79f727a6c1d51e83381c3fd1d8e1fddeff7de99fe70ca06bc1bff49", "submitted_at": "2025-12-01T01:18:21Z", "updated_at": "2025-12-01T01:18:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.01174v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.01174", "catalog_id": "arxiv:2512.01174", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-01", "identifiers": { "arxiv": "2512.01174" }, "record_updated_at": "2025-12-01" }, "genre": "preprint", "id": "arxiv:2512.01174", "issued": { "date-parts": [ [ 2025, 12, 1 ] ] }, "note": "First published: 2025-12-01; Recorded arXiv version: v1; Recorded version updated: 2025-12-01T01:18:21Z.", "title": "DrawingBench: Evaluating Spatial Reasoning and UI Interaction Capabilities of Large Language Models through Mouse-Based Drawing Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.01979v1", "archive": "arXiv", "archive_location": "2512.01979", "author": [ { "literal": "Aiden Yiliu Li" }, { "literal": "Bizhi Yu" }, { "literal": "Daoan Lei" }, { "literal": "Tianhe Ren" }, { "literal": "Shilong Liu" } ], "citation-key": "arxiv_2512_01979", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4868c3c85545532e9c208623596d3810c85e32e3cf11429a68847ac11c2ba1cb", "submitted_at": "2025-12-01T18:37:19Z", "updated_at": "2025-12-01T18:37:19Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.01979v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.01979", "catalog_id": "arxiv:2512.01979", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-01", "identifiers": { "arxiv": "2512.01979" }, "record_updated_at": "2025-12-01" }, "genre": "preprint", "id": "arxiv:2512.01979", "issued": { "date-parts": [ [ 2025, 12, 1 ] ] }, "note": "First published: 2025-12-01; Recorded arXiv version: v1; Recorded version updated: 2025-12-01T18:37:19Z.", "title": "Chain-of-Ground: Improving GUI Grounding via Iterative Reasoning and Reference Feedback", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.02423v1", "archive": "arXiv", "archive_location": "2512.02423", "author": [ { "literal": "Haolong Yan" }, { "literal": "Yeqing Shen" }, { "literal": "Xin Huang" }, { "literal": "Jia Wang" }, { "literal": "Kaijun Tan" }, { "literal": "Zhixuan Liang" }, { "literal": "Hongxin Li" }, { "literal": "Zheng Ge" }, { "literal": "Osamu Yoshie" }, { "literal": "Si Li" }, { "literal": "Xiangyu Zhang" }, { "literal": "Daxin Jiang" } ], "citation-key": "arxiv_2512_02423", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fda844ac6767e6d9dfd16919042469e2070ec7f092e0b01fb5a243ed6eb4c282", "submitted_at": "2025-12-02T05:11:23Z", "updated_at": "2025-12-02T05:11:23Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.02423v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.02423", "catalog_id": "arxiv:2512.02423", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-02", "identifiers": { "arxiv": "2512.02423" }, "record_updated_at": "2025-12-02" }, "genre": "preprint", "id": "arxiv:2512.02423", "issued": { "date-parts": [ [ 2025, 12, 2 ] ] }, "note": "First published: 2025-12-02; Recorded arXiv version: v1; Recorded version updated: 2025-12-02T05:11:23Z.", "title": "GUI Exploration Lab: Enhancing Screen Navigation in Agents via Multi-Turn Reinforcement Learning", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.04105v1", "archive": "arXiv", "archive_location": "2512.04105", "author": [ { "literal": "Jinzhe Tan" }, { "literal": "Karim Benyekhlef" } ], "citation-key": "arxiv_2512_04105", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d362a839fd1ee917ebdde6fcd794640523e930cd711769154532200903fe7de5", "submitted_at": "2025-11-28T07:26:29Z", "updated_at": "2025-11-28T07:26:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.04105v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.04105", "catalog_id": "arxiv:2512.04105", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-11-28", "identifiers": { "arxiv": "2512.04105" }, "record_updated_at": "2025-11-28" }, "genre": "preprint", "id": "arxiv:2512.04105", "issued": { "date-parts": [ [ 2025, 11, 28 ] ] }, "note": "First published: 2025-11-28; Recorded arXiv version: v1; Recorded version updated: 2025-11-28T07:26:29Z.", "title": "LegalWebAgent: Empowering Access to Justice via LLM-Based Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.05941v1", "archive": "arXiv", "archive_location": "2512.05941", "author": [ { "literal": "Zhiyuan Jiang" }, { "literal": "Shenghao Xie" }, { "literal": "Wenyi Li" }, { "literal": "Wenqiang Zu" }, { "literal": "Peihang Li" }, { "literal": "Jiahao Qiu" }, { "literal": "Siqi Pei" }, { "literal": "Lei Ma" }, { "literal": "Tiejun Huang" }, { "literal": "Mengdi Wang" }, { "literal": "Shilong Liu" } ], "citation-key": "arxiv_2512_05941", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bac971bd0e7cd946f81e28d5976eb933cbb95b97ca846ee624a0c36b48760de7", "submitted_at": "2025-12-05T18:39:12Z", "updated_at": "2025-12-05T18:39:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.05941v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.05941", "catalog_id": "arxiv:2512.05941", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-05", "identifiers": { "arxiv": "2512.05941" }, "record_updated_at": "2025-12-05" }, "genre": "preprint", "id": "arxiv:2512.05941", "issued": { "date-parts": [ [ 2025, 12, 5 ] ] }, "note": "First published: 2025-12-05; Recorded arXiv version: v1; Recorded version updated: 2025-12-05T18:39:12Z.", "title": "Zoom in, Click out: Unlocking and Evaluating the Potential of Zooming for GUI Grounding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.07725v1", "archive": "arXiv", "archive_location": "2512.07725", "author": [ { "literal": "Alisha Ukani" }, { "literal": "Hamed Haddadi" }, { "literal": "Ali Shahin Shamsabadi" }, { "literal": "Peter Snyder" } ], "citation-key": "arxiv_2512_07725", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2576701bbdc4fd8f1f571ed691b06185d1e16bf4955a5c99b10200d796ba6525", "submitted_at": "2025-12-08T17:16:12Z", "updated_at": "2025-12-08T17:16:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.07725v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.07725", "catalog_id": "arxiv:2512.07725", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-08", "identifiers": { "arxiv": "2512.07725" }, "record_updated_at": "2025-12-08" }, "genre": "preprint", "id": "arxiv:2512.07725", "issued": { "date-parts": [ [ 2025, 12, 8 ] ] }, "note": "First published: 2025-12-08; Recorded arXiv version: v1; Recorded version updated: 2025-12-08T17:16:12Z.", "title": "Privacy Practices of Browser Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.11611v1", "archive": "arXiv", "archive_location": "2512.11611", "author": [ { "literal": "Chunyi Li" }, { "literal": "Longfei Li" }, { "literal": "Zicheng Zhang" }, { "literal": "Xiaohong Liu" }, { "literal": "Min Tang" }, { "literal": "Weisi Lin" }, { "literal": "Guangtao Zhai" } ], "citation-key": "arxiv_2512_11611", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "484a51656db4466d5e8630d64fa78d3cf10043568488f5a2e9bae679fa800cbe", "submitted_at": "2025-12-12T14:49:32Z", "updated_at": "2025-12-12T14:49:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.11611v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.11611", "catalog_id": "arxiv:2512.11611", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-12", "identifiers": { "arxiv": "2512.11611" }, "record_updated_at": "2025-12-12" }, "genre": "preprint", "id": "arxiv:2512.11611", "issued": { "date-parts": [ [ 2025, 12, 12 ] ] }, "note": "First published: 2025-12-12; Recorded arXiv version: v1; Recorded version updated: 2025-12-12T14:49:32Z.", "title": "Using GUI Agent for Electronic Design Automation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.12594v2", "archive": "arXiv", "archive_location": "2512.12594", "author": [ { "literal": "Luoxi Meng" }, { "literal": "Henry Feng" }, { "literal": "Ilia Shumailov" }, { "literal": "Earlence Fernandes" } ], "citation-key": "arxiv_2512_12594", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fa6b4d9fa34610ec6fa8c0ac8b075127365aad57a20d814a2e838a223cf8bd56", "submitted_at": "2025-12-14T08:25:31Z", "updated_at": "2026-03-04T01:49:57Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2512.12594v2" }, "canonical_source_url": "https://arxiv.org/abs/2512.12594", "catalog_id": "arxiv:2512.12594", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-14", "identifiers": { "arxiv": "2512.12594" }, "record_updated_at": "2026-03-04" }, "genre": "preprint", "id": "arxiv:2512.12594", "issued": { "date-parts": [ [ 2025, 12, 14 ] ] }, "note": "First published: 2025-12-14; Recorded arXiv version: v2; Recorded version updated: 2026-03-04T01:49:57Z.", "title": "ceLLMate: Sandboxing Browser AI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2512.12634v4", "archive": "arXiv", "archive_location": "2512.12634", "author": [ { "literal": "Youngmin Im" }, { "literal": "Byeongung Jo" }, { "literal": "Jaeyoung Wi" }, { "literal": "Seungwoo Baek" }, { "literal": "Tae Hoon Min" }, { "literal": "Joo Hyung Lee" }, { "literal": "Sangeun Oh" }, { "literal": "Insik Shin" }, { "literal": "Sunjae Lee" } ], "citation-key": "arxiv_2512_12634", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5982901598e43dcadf63b4a950b2fff68e85be71d0f483648b535d82d7745bd7", "submitted_at": "2025-12-14T10:41:39Z", "updated_at": "2026-05-31T23:48:10Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2512.12634v4" }, "canonical_source_url": "https://arxiv.org/abs/2512.12634", "catalog_id": "arxiv:2512.12634", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-14", "identifiers": { "arxiv": "2512.12634" }, "record_updated_at": "2026-05-31" }, "genre": "preprint", "id": "arxiv:2512.12634", "issued": { "date-parts": [ [ 2025, 12, 14 ] ] }, "note": "First published: 2025-12-14; Recorded arXiv version: v4; Recorded version updated: 2026-05-31T23:48:10Z.", "title": "MobiBench: Multi-Branch, Modular Benchmark for Mobile GUI Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2512.14014v2", "archive": "arXiv", "archive_location": "2512.14014", "author": [ { "literal": "Shufan Li" }, { "literal": "Konstantinos Kallidromitis" }, { "literal": "Akash Gokul" }, { "literal": "Yusuke Kato" }, { "literal": "Kazuki Kozuka" }, { "literal": "Aditya Grover" } ], "citation-key": "arxiv_2512_14014", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e6228822cf64c6aeb1836bee52a6515123bd0d44d77cceb23e785bb4ff9aef47", "submitted_at": "2025-12-16T02:16:42Z", "updated_at": "2026-10-07T23:06:49Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2512.14014v2" }, "canonical_source_url": "https://arxiv.org/abs/2512.14014", "catalog_id": "arxiv:2512.14014", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-16", "identifiers": { "arxiv": "2512.14014" }, "record_updated_at": "2026-10-07" }, "genre": "preprint", "id": "arxiv:2512.14014", "issued": { "date-parts": [ [ 2025, 12, 16 ] ] }, "note": "First published: 2025-12-16; Recorded arXiv version: v2; Recorded version updated: 2026-10-07T23:06:49Z.", "title": "MobileWorldBench: Towards Semantic World Modeling For Mobile Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2512.15431v2", "archive": "arXiv", "archive_location": "2512.15431", "author": [ { "literal": "Haolong Yan" }, { "literal": "Jia Wang" }, { "literal": "Xin Huang" }, { "literal": "Yeqing Shen" }, { "literal": "Ziyang Meng" }, { "literal": "Zhimin Fan" }, { "literal": "Kaijun Tan" }, { "literal": "Jin Gao" }, { "literal": "Lieyu Shi" }, { "literal": "Mi Yang" }, { "literal": "Shiliang Yang" }, { "literal": "Zhirui Wang" }, { "literal": "Brian Li" }, { "literal": "Kang An" }, { "literal": "Chenyang Li" }, { "literal": "Lei Lei" }, { "literal": "Mengmeng Duan" }, { "literal": "Danxun Liang" }, { "literal": "Guodong Liu" }, { "literal": "Hang Cheng" }, { "literal": "Hao Wu" }, { "literal": "Jie Dong" }, { "literal": "Junhao Huang" }, { "literal": "Mei Chen" }, { "literal": "Renjie Yu" }, { "literal": "Shunshan Li" }, { "literal": "Xu Zhou" }, { "literal": "Yiting Dai" }, { "literal": "Yineng Deng" }, { "literal": "Yingdan Liang" }, { "literal": "Zelin Chen" }, { "literal": "Wen Sun" }, { "literal": "Chengxu Yan" }, { "literal": "Chunqin Xu" }, { "literal": "Dong Li" }, { "literal": "Fengqiong Xiao" }, { "literal": "Guanghao Fan" }, { "literal": "Guopeng Li" }, { "literal": "Guozhen Peng" }, { "literal": "Hongbing Li" }, { "literal": "Hang Li" }, { "literal": "Hongming Chen" }, { "literal": "Jingjing Xie" }, { "literal": "Jianyong Li" }, { "literal": "Jingyang Zhang" }, { "literal": "Jiaju Ren" }, { "literal": "Jiayu Yuan" }, { "literal": "Jianpeng Yin" }, { "literal": "Kai Cao" }, { "literal": "Liang Zhao" }, { "literal": "Liguo Tan" }, { "literal": "Liying Shi" }, { "literal": "Mengqiang Ren" }, { "literal": "Min Xu" }, { "literal": "Manjiao Liu" }, { "literal": "Mao Luo" }, { "literal": "Mingxin Wan" }, { "literal": "Na Wang" }, { "literal": "Nan Wu" }, { "literal": "Ning Wang" }, { "literal": "Peiyao Ma" }, { "literal": "Qingzhou Zhang" }, { "literal": "Qiao Wang" }, { "literal": "Qinlin Zeng" }, { "literal": "Qiong Gao" }, { "literal": "Qiongyao Li" }, { "literal": "Shangwu Zhong" }, { "literal": "Shuli Gao" }, { "literal": "Shaofan Liu" }, { "literal": "Shisi Gao" }, { "literal": "Shuang Luo" }, { "literal": "Xingbin Liu" }, { "literal": "Xiaojia Liu" }, { "literal": "Xiaojie Hou" }, { "literal": "Xin Liu" }, { "literal": "Xuanti Feng" }, { "literal": "Xuedan Cai" }, { "literal": "Xuan Wen" }, { "literal": "Xianwei Zhu" }, { "literal": "Xin Liang" }, { "literal": "Xin Liu" }, { "literal": "Xin Zhou" }, { "literal": "Yifan Sui" }, { "literal": "Yingxiu Zhao" }, { "literal": "Yukang Shi" }, { "literal": "Yunfang Xu" }, { "literal": "Yuqing Zeng" }, { "literal": "Yixun Zhang" }, { "literal": "Zejia Weng" }, { "literal": "Zhonghao Yan" }, { "literal": "Zhiguo Huang" }, { "literal": "Zhuoyu Wang" }, { "literal": "Zihan Yan" }, { "literal": "Zheng Ge" }, { "literal": "Jing Li" }, { "literal": "Yibo Zhu" }, { "literal": "Binxing Jiao" }, { "literal": "Xiangyu Zhang" }, { "literal": "Daxin Jiang" } ], "citation-key": "arxiv_2512_15431", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2fc824bdeabca3451fd895447a0321c507da1ac7bc7cc24184c7768992549daf", "submitted_at": "2025-12-17T13:26:30Z", "updated_at": "2025-12-19T17:36:21Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2512.15431v2" }, "canonical_source_url": "https://arxiv.org/abs/2512.15431", "catalog_id": "arxiv:2512.15431", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-17", "identifiers": { "arxiv": "2512.15431" }, "record_updated_at": "2025-12-19" }, "genre": "preprint", "id": "arxiv:2512.15431", "issued": { "date-parts": [ [ 2025, 12, 17 ] ] }, "note": "First published: 2025-12-17; Recorded arXiv version: v2; Recorded version updated: 2025-12-19T17:36:21Z.", "title": "Step-GUI Technical Report", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2512.16295v1", "archive": "arXiv", "archive_location": "2512.16295", "author": [ { "literal": "Zhenyu Wu" }, { "literal": "Jingjing Xie" }, { "literal": "Zehao Li" }, { "literal": "Bowen Yang" }, { "literal": "Qiushi Sun" }, { "literal": "Zhaoyang Liu" }, { "literal": "Zhoumianze Liu" }, { "literal": "Yu Qiao" }, { "literal": "Xiangyu Yue" }, { "literal": "Zun Wang" }, { "literal": "Zichen Ding" } ], "citation-key": "arxiv_2512_16295", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "67a1bd40b117b34e5fb994b1c2c93cd285714097d1bc9730baeeaba6c31a7b8a", "submitted_at": "2025-12-18T08:29:50Z", "updated_at": "2025-12-18T08:29:50Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.16295v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.16295", "catalog_id": "arxiv:2512.16295", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-18", "identifiers": { "arxiv": "2512.16295" }, "record_updated_at": "2025-12-18" }, "genre": "preprint", "id": "arxiv:2512.16295", "issued": { "date-parts": [ [ 2025, 12, 18 ] ] }, "note": "First published: 2025-12-18; Recorded arXiv version: v1; Recorded version updated: 2025-12-18T08:29:50Z.", "title": "OS-Oracle: A Comprehensive Framework for Cross-Platform GUI Critic Models", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.16501v1", "archive": "arXiv", "archive_location": "2512.16501", "author": [ { "literal": "Beitong Zhou" }, { "literal": "Zhexiao Huang" }, { "literal": "Yuan Guo" }, { "literal": "Zhangxuan Gu" }, { "literal": "Tianyu Xia" }, { "literal": "Zichen Luo" }, { "literal": "Fei Tang" }, { "literal": "Dehan Kong" }, { "literal": "Yanyi Shang" }, { "literal": "Suling Ou" }, { "literal": "Zhenlin Guo" }, { "literal": "Changhua Meng" }, { "literal": "Shuheng Shen" } ], "citation-key": "arxiv_2512_16501", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4bc8a1da3993ead9b36af47a9332fe9a1a2427101ad0d91da654b4a4b25141df", "submitted_at": "2025-12-18T13:09:09Z", "updated_at": "2025-12-18T13:09:09Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.16501v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.16501", "catalog_id": "arxiv:2512.16501", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-18", "identifiers": { "arxiv": "2512.16501" }, "record_updated_at": "2025-12-18" }, "genre": "preprint", "id": "arxiv:2512.16501", "issued": { "date-parts": [ [ 2025, 12, 18 ] ] }, "note": "First published: 2025-12-18; Recorded arXiv version: v1; Recorded version updated: 2025-12-18T13:09:09Z.", "title": "VenusBench-GD: A Comprehensive Multi-Platform GUI Benchmark for Diverse Grounding Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.17221v3", "archive": "arXiv", "archive_location": "2512.17221", "author": [ { "literal": "Brandon Huang" }, { "literal": "Hang Hua" }, { "literal": "Zhuoran Yu" }, { "literal": "Trevor Darrell" }, { "literal": "Rogerio Feris" }, { "literal": "Roei Herzig" } ], "citation-key": "arxiv_2512_17221", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "37cd0aa99ed8bad167e25f6e2b48e7cd1cf01249508555f657215213ee94a301", "submitted_at": "2025-12-19T04:09:24Z", "updated_at": "2025-12-31T17:30:11Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2512.17221v3" }, "canonical_source_url": "https://arxiv.org/abs/2512.17221", "catalog_id": "arxiv:2512.17221", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-19", "identifiers": { "arxiv": "2512.17221" }, "record_updated_at": "2025-12-31" }, "genre": "preprint", "id": "arxiv:2512.17221", "issued": { "date-parts": [ [ 2025, 12, 19 ] ] }, "note": "First published: 2025-12-19; Recorded arXiv version: v3; Recorded version updated: 2025-12-31T17:30:11Z.", "title": "DAVE: A VLM Vision Encoder for Document Understanding and Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2512.19432v3", "archive": "arXiv", "archive_location": "2512.19432", "author": [ { "literal": "Quyu Kong" }, { "literal": "Xu Zhang" }, { "literal": "Zhenyu Yang" }, { "literal": "Nolan Gao" }, { "literal": "Chen Liu" }, { "literal": "Panrong Tong" }, { "literal": "Chenglin Cai" }, { "literal": "Hanzhang Zhou" }, { "literal": "Jianan Zhang" }, { "literal": "Liangyu Chen" }, { "literal": "Zhidan Liu" }, { "literal": "Steven Hoi" }, { "literal": "Yue Wang" } ], "citation-key": "arxiv_2512_19432", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cf3258fd2286b0ee6babb69b9bfc459ecf40536eec498a3af3403d2971791f3d", "submitted_at": "2025-12-22T14:31:28Z", "updated_at": "2025-12-30T07:15:15Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2512.19432v3" }, "canonical_source_url": "https://arxiv.org/abs/2512.19432", "catalog_id": "arxiv:2512.19432", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-22", "identifiers": { "arxiv": "2512.19432" }, "record_updated_at": "2025-12-30" }, "genre": "preprint", "id": "arxiv:2512.19432", "issued": { "date-parts": [ [ 2025, 12, 22 ] ] }, "note": "First published: 2025-12-22; Recorded arXiv version: v3; Recorded version updated: 2025-12-30T07:15:15Z.", "title": "MobileWorld: Benchmarking Autonomous Mobile Agents in Agent-User Interactive and MCP-Augmented Environments", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2512.21302v1", "archive": "arXiv", "archive_location": "2512.21302", "author": [ { "literal": "Yue Cao" }, { "literal": "Yingyao Wang" }, { "literal": "Pi Bu" }, { "literal": "Jingxuan Xing" }, { "literal": "Wei Jiang" }, { "literal": "Zekun Zhu" }, { "literal": "Junpeng Ma" }, { "literal": "Sashuai Zhou" }, { "literal": "Tong Lu" }, { "literal": "Jun Song" }, { "literal": "Yu Cheng" }, { "literal": "Yuning Jiang" }, { "literal": "Bo Zheng" } ], "citation-key": "arxiv_2512_21302", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "597935923ce9e59720771f3d92e921f2842f908ad81b3ab31a87ed9d9f5c5774", "submitted_at": "2025-12-24T17:40:42Z", "updated_at": "2025-12-24T17:40:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.21302v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.21302", "catalog_id": "arxiv:2512.21302", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-24", "identifiers": { "arxiv": "2512.21302" }, "record_updated_at": "2025-12-24" }, "genre": "preprint", "id": "arxiv:2512.21302", "issued": { "date-parts": [ [ 2025, 12, 24 ] ] }, "note": "First published: 2025-12-24; Recorded arXiv version: v1; Recorded version updated: 2025-12-24T17:40:42Z.", "title": "AndroidLens: Long-latency Evaluation with Nested Sub-targets for Android GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2512.23128v3", "archive": "arXiv", "archive_location": "2512.23128", "author": [ { "literal": "Karolina Korgul" }, { "literal": "Yushi Yang" }, { "literal": "Arkadiusz Drohomirecki" }, { "literal": "Piotr Błaszczyk" }, { "literal": "Will Howard" }, { "literal": "Lukas Aichberger" }, { "literal": "Chris Russell" }, { "literal": "Philip H. S. Torr" }, { "literal": "Adam Mahdi" }, { "literal": "Adel Bibi" } ], "citation-key": "arxiv_2512_23128", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a255cccbe3f16f498918126724c391911e1d745c093189790cbab715fd28a2d2", "submitted_at": "2025-12-29T01:09:10Z", "updated_at": "2026-07-02T16:12:04Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2512.23128v3" }, "canonical_source_url": "https://arxiv.org/abs/2512.23128", "catalog_id": "arxiv:2512.23128", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-29", "identifiers": { "arxiv": "2512.23128" }, "record_updated_at": "2026-07-02" }, "genre": "preprint", "id": "arxiv:2512.23128", "issued": { "date-parts": [ [ 2025, 12, 29 ] ] }, "note": "First published: 2025-12-29; Recorded arXiv version: v3; Recorded version updated: 2026-07-02T16:12:04Z.", "title": "It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2512.24965v1", "archive": "arXiv", "archive_location": "2512.24965", "author": [ { "literal": "Siyuan Hu" }, { "literal": "Kevin Qinghong Lin" }, { "literal": "Mike Zheng Shou" } ], "citation-key": "arxiv_2512_24965", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b7d1f358b450f2a6c3deb00f1b16c54d443f48fda3a7dc70c77456bebf35ad7c", "submitted_at": "2025-12-31T16:51:14Z", "updated_at": "2025-12-31T16:51:14Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2512.24965v1" }, "canonical_source_url": "https://arxiv.org/abs/2512.24965", "catalog_id": "arxiv:2512.24965", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-31", "identifiers": { "arxiv": "2512.24965" }, "record_updated_at": "2025-12-31" }, "genre": "preprint", "id": "arxiv:2512.24965", "issued": { "date-parts": [ [ 2025, 12, 31 ] ] }, "note": "First published: 2025-12-31; Recorded arXiv version: v1; Recorded version updated: 2025-12-31T16:51:14Z.", "title": "ShowUI-$π$: Flow-based Generative Models as GUI Dexterous Hands", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.02399v1", "archive": "arXiv", "archive_location": "2601.02399", "author": [ { "literal": "Jiaxin Ai" }, { "literal": "Yukang Feng" }, { "literal": "Fanrui Zhang" }, { "literal": "Jianwen Sun" }, { "literal": "Zizhen Li" }, { "literal": "Chuanhao Li" }, { "literal": "Yifan Chang" }, { "literal": "Wenxiao Wu" }, { "literal": "Ruoxi Wang" }, { "literal": "Mingliang Zhai" }, { "literal": "Kaipeng Zhang" } ], "citation-key": "arxiv_2601_02399", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "afab3dcd2444bf8333f2a2fea887819f6fca55c41209d8c449f6211d1c62b819", "submitted_at": "2025-12-30T01:49:46Z", "updated_at": "2025-12-30T01:49:46Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.02399v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.02399", "catalog_id": "arxiv:2601.02399", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2025-12-30", "identifiers": { "arxiv": "2601.02399" }, "record_updated_at": "2025-12-30" }, "genre": "preprint", "id": "arxiv:2601.02399", "issued": { "date-parts": [ [ 2025, 12, 30 ] ] }, "note": "First published: 2025-12-30; Recorded arXiv version: v1; Recorded version updated: 2025-12-30T01:49:46Z.", "title": "ProSoftArena: Benchmarking Hierarchical Capabilities of Multimodal Agents in Professional Software Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.02439v6", "archive": "arXiv", "archive_location": "2601.02439", "author": [ { "literal": "Hao Bai" }, { "literal": "Alexey Taymanov" }, { "literal": "Tong Zhang" }, { "literal": "Aviral Kumar" }, { "literal": "Spencer Whitehead" } ], "citation-key": "arxiv_2601_02439", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ae8751c4fae3b4f838eba59ffa8be37a1a5f4fa497dfd69fa3075f8581547f51", "submitted_at": "2026-01-05T09:35:11Z", "updated_at": "2026-05-02T18:45:09Z", "version": 6, "versioned_url": "https://arxiv.org/abs/2601.02439v6" }, "canonical_source_url": "https://arxiv.org/abs/2601.02439", "catalog_id": "arxiv:2601.02439", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-05", "identifiers": { "arxiv": "2601.02439" }, "record_updated_at": "2026-05-02" }, "genre": "preprint", "id": "arxiv:2601.02439", "issued": { "date-parts": [ [ 2026, 1, 5 ] ] }, "note": "First published: 2026-01-05; Recorded arXiv version: v6; Recorded version updated: 2026-05-02T18:45:09Z.", "title": "WebGym: Scaling Training Environments for Visual Web Agents with Realistic Tasks", "type": "article", "version": "6" }, { "URL": "https://arxiv.org/abs/2601.04126v3", "archive": "arXiv", "archive_location": "2601.04126", "author": [ { "literal": "Ziyun Zhang" }, { "literal": "Zezhou Wang" }, { "literal": "Xiaoyi Zhang" }, { "literal": "Zongyu Guo" }, { "literal": "Jiahao Li" }, { "literal": "Bin Li" }, { "literal": "Yan Lu" } ], "citation-key": "arxiv_2601_04126", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d4d0a59f1eb8197b4d389f5f40957fb8aeb1e22d0739f9877b26d57ed70fe59b", "submitted_at": "2026-01-07T17:40:08Z", "updated_at": "2026-06-30T13:54:32Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2601.04126v3" }, "canonical_source_url": "https://arxiv.org/abs/2601.04126", "catalog_id": "arxiv:2601.04126", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-07", "identifiers": { "arxiv": "2601.04126" }, "record_updated_at": "2026-06-30" }, "genre": "preprint", "id": "arxiv:2601.04126", "issued": { "date-parts": [ [ 2026, 1, 7 ] ] }, "note": "First published: 2026-01-07; Recorded arXiv version: v3; Recorded version updated: 2026-06-30T13:54:32Z.", "title": "InfiniteWeb: Scalable Web Environment Synthesis for GUI Agent Training", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2601.04500v1", "archive": "arXiv", "archive_location": "2601.04500", "author": [ { "literal": "Yifei Gao" }, { "literal": "Jiang Wu" }, { "literal": "Xiaoyi Chen" }, { "literal": "Yifan Yang" }, { "literal": "Zhe Cui" }, { "literal": "Tianyi Ma" }, { "literal": "Jiaming Zhang" }, { "literal": "Jitao Sang" } ], "citation-key": "arxiv_2601_04500", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "47a340b19cf399f2ae7851482030d85e572b0f6ad8415c67657beec737a80e06", "submitted_at": "2026-01-08T02:07:53Z", "updated_at": "2026-01-08T02:07:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.04500v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.04500", "catalog_id": "arxiv:2601.04500", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-08", "identifiers": { "arxiv": "2601.04500" }, "record_updated_at": "2026-01-08" }, "genre": "preprint", "id": "arxiv:2601.04500", "issued": { "date-parts": [ [ 2026, 1, 8 ] ] }, "note": "First published: 2026-01-08; Recorded arXiv version: v1; Recorded version updated: 2026-01-08T02:07:53Z.", "title": "GUITester: Enabling GUI Agents for Exploratory Defect Discovery", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.05787v2", "archive": "arXiv", "archive_location": "2601.05787", "author": [ { "literal": "Zezhou Wang" }, { "literal": "Ziyun Zhang" }, { "literal": "Xiaoyi Zhang" }, { "literal": "Zhuzhong Qian" }, { "literal": "Yan Lu" } ], "citation-key": "arxiv_2601_05787", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e342723078b50e612df749df80eadfd16e0fe29f5875777c1b3189d8b5981512", "submitted_at": "2026-01-09T13:26:38Z", "updated_at": "2026-02-10T15:12:17Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.05787v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.05787", "catalog_id": "arxiv:2601.05787", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-09", "identifiers": { "arxiv": "2601.05787" }, "record_updated_at": "2026-02-10" }, "genre": "preprint", "id": "arxiv:2601.05787", "issued": { "date-parts": [ [ 2026, 1, 9 ] ] }, "note": "First published: 2026-01-09; Recorded arXiv version: v2; Recorded version updated: 2026-02-10T15:12:17Z.", "title": "From Off-Policy to On-Policy: Enhancing GUI Agents via Bi-level Expert-to-Policy Assimilation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.08406v1", "archive": "arXiv", "archive_location": "2601.08406", "author": [ { "literal": "Xinyi Wu" }, { "literal": "Jiagui Chen" }, { "literal": "Geng Hong" }, { "literal": "Jiayi Dong" }, { "literal": "Xudong Pan" }, { "literal": "Jiarun Dai" }, { "literal": "Min Yang" } ], "citation-key": "arxiv_2601_08406", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "626702d22786336ff65b6e89d97e2034ff7aa144fab4ecc286d169fca6993410", "submitted_at": "2026-01-13T10:21:28Z", "updated_at": "2026-01-13T10:21:28Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.08406v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.08406", "catalog_id": "arxiv:2601.08406", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-13", "identifiers": { "arxiv": "2601.08406" }, "record_updated_at": "2026-01-13" }, "genre": "preprint", "id": "arxiv:2601.08406", "issued": { "date-parts": [ [ 2026, 1, 13 ] ] }, "note": "First published: 2026-01-13; Recorded arXiv version: v1; Recorded version updated: 2026-01-13T10:21:28Z.", "title": "WebTrap Park: An Automated Platform for Systematic Security Evaluation of Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.08605v2", "archive": "arXiv", "archive_location": "2601.08605", "author": [ { "literal": "Wenyuan Zhang" }, { "literal": "Xinghua Zhang" }, { "literal": "Haiyang Yu" }, { "literal": "Shuaiyi Nie" }, { "literal": "Bingli Wu" }, { "literal": "Juwei Yue" }, { "literal": "Tingwen Liu" }, { "literal": "Yongbin Li" } ], "citation-key": "arxiv_2601_08605", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "73df8c5b417c3387530b7730be3aed8b69374c06fdd3573f56147b19359c6912", "submitted_at": "2026-01-13T14:48:34Z", "updated_at": "2026-04-15T14:59:17Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.08605v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.08605", "catalog_id": "arxiv:2601.08605", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-13", "identifiers": { "arxiv": "2601.08605" }, "record_updated_at": "2026-04-15" }, "genre": "preprint", "id": "arxiv:2601.08605", "issued": { "date-parts": [ [ 2026, 1, 13 ] ] }, "note": "First published: 2026-01-13; Recorded arXiv version: v2; Recorded version updated: 2026-04-15T14:59:17Z.", "title": "ExpSeek: Self-Triggered Experience Seeking for Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.09636v2", "archive": "arXiv", "archive_location": "2601.09636", "author": [ { "literal": "Yibo Lyu" }, { "literal": "Gongwei Chen" }, { "literal": "Rui Shao" }, { "literal": "Weili Guan" }, { "literal": "Liqiang Nie" } ], "citation-key": "arxiv_2601_09636", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d8b697b91942291403491581a61c135c6bfc8b0f3c2a9b9bab0f926af7208fce", "submitted_at": "2026-01-14T17:12:48Z", "updated_at": "2026-05-13T02:54:19Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.09636v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.09636", "catalog_id": "arxiv:2601.09636", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-14", "identifiers": { "arxiv": "2601.09636" }, "record_updated_at": "2026-05-13" }, "genre": "preprint", "id": "arxiv:2601.09636", "issued": { "date-parts": [ [ 2026, 1, 14 ] ] }, "note": "First published: 2026-01-14; Recorded arXiv version: v2; Recorded version updated: 2026-05-13T02:54:19Z.", "title": "PersonalAlign: Hierarchical Implicit Intent Alignment for Personalized GUI Agent with Long-Term User-Centric Records", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.12349v3", "archive": "arXiv", "archive_location": "2601.12349", "author": [ { "literal": "Yi Qian" }, { "literal": "Kunwei Qian" }, { "literal": "Xingbang He" }, { "literal": "Ligeng Chen" }, { "literal": "Jikang Zhang" }, { "literal": "Tiantai Zhang" }, { "literal": "Haiyang Wei" }, { "literal": "Linzhang Wang" }, { "literal": "Hao Wu" }, { "literal": "Bing Mao" } ], "citation-key": "arxiv_2601_12349", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "21848dea9e000125550fff7fe3844a57bf100215bf9eb800cbdbb612675f7074", "submitted_at": "2026-01-18T10:54:54Z", "updated_at": "2026-07-15T15:32:11Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2601.12349v3" }, "canonical_source_url": "https://arxiv.org/abs/2601.12349", "catalog_id": "arxiv:2601.12349", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-18", "identifiers": { "arxiv": "2601.12349" }, "record_updated_at": "2026-07-15" }, "genre": "preprint", "id": "arxiv:2601.12349", "issued": { "date-parts": [ [ 2026, 1, 18 ] ] }, "note": "First published: 2026-01-18; Recorded arXiv version: v3; Recorded version updated: 2026-07-15T15:32:11Z.", "title": "Mind the Gap: Action Rebinding Attacks against Android GUI Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2601.13060v1", "archive": "arXiv", "archive_location": "2601.13060", "author": [ { "literal": "Zecheng Li" }, { "literal": "Zhihui Cao" }, { "literal": "Wenke Huang" }, { "literal": "Yudong Zhang" }, { "literal": "Keying Qi" }, { "literal": "Rui Wang" }, { "literal": "Zeyu Zheng" }, { "literal": "Jian Zhao" }, { "literal": "Hao Zhu" }, { "literal": "Hengxin Wu" }, { "literal": "Yuran Wang" }, { "literal": "Guitao Fan" }, { "literal": "Guokun Wu" }, { "literal": "Yicong Liu" }, { "literal": "Zhilin Gao" }, { "literal": "Haikun Xu" }, { "literal": "He Yang" }, { "literal": "Minqi Xiang" }, { "literal": "Xingyu Liu" }, { "literal": "Zuojian Wang" } ], "citation-key": "arxiv_2601_13060", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "aa8cb4137e862bf8031a55bc8f81e1d6c903733f2cfac9cf758e5f2f92884481", "submitted_at": "2026-01-19T13:50:43Z", "updated_at": "2026-01-19T13:50:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.13060v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.13060", "catalog_id": "arxiv:2601.13060", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-19", "identifiers": { "arxiv": "2601.13060" }, "record_updated_at": "2026-01-19" }, "genre": "preprint", "id": "arxiv:2601.13060", "issued": { "date-parts": [ [ 2026, 1, 19 ] ] }, "note": "First published: 2026-01-19; Recorded arXiv version: v1; Recorded version updated: 2026-01-19T13:50:43Z.", "title": "MagicGUI-RMS: A Multi-Agent Reward Model System for Self-Evolving GUI Agents via Automated Feedback Reflux", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.16356v1", "archive": "arXiv", "archive_location": "2601.16356", "author": [ { "literal": "Simret Araya Gebreegziabher" }, { "literal": "Yukun Yang" }, { "literal": "Charles Chiang" }, { "literal": "Hojun Yoo" }, { "literal": "Chaoran Chen" }, { "literal": "Hyo Jin Do" }, { "literal": "Zahra Ashktorab" }, { "literal": "Werner Geyer" }, { "literal": "Diego Gómez-Zará" }, { "literal": "Toby Jia-Jun Li" } ], "citation-key": "arxiv_2601_16356", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1bdc29648446cd79e0c0b9f72814457b55258d501283ce2fa2dedcd8262f885b", "submitted_at": "2026-01-22T22:42:40Z", "updated_at": "2026-01-22T22:42:40Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.16356v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.16356", "catalog_id": "arxiv:2601.16356", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-22", "identifiers": { "arxiv": "2601.16356" }, "record_updated_at": "2026-01-22" }, "genre": "preprint", "id": "arxiv:2601.16356", "issued": { "date-parts": [ [ 2026, 1, 22 ] ] }, "note": "First published: 2026-01-22; Recorded arXiv version: v1; Recorded version updated: 2026-01-22T22:42:40Z.", "title": "The Behavioral Fabric of LLM-Powered GUI Agents: Human Values and Interaction Outcomes", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.17722v1", "archive": "arXiv", "archive_location": "2601.17722", "author": [ { "literal": "Ying Mo" }, { "literal": "Yu Bai" }, { "literal": "Dapeng Sun" }, { "literal": "Yuqian Shi" }, { "literal": "Yukai Miao" }, { "literal": "Li Chen" }, { "literal": "Dan Li" } ], "citation-key": "arxiv_2601_17722", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fa7c8ce1b0e584c5279691ff06cd522d7c8ea1f4801d7074324c6845ce482efb", "submitted_at": "2026-01-25T06:58:15Z", "updated_at": "2026-01-25T06:58:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.17722v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.17722", "catalog_id": "arxiv:2601.17722", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-25", "identifiers": { "arxiv": "2601.17722" }, "record_updated_at": "2026-01-25" }, "genre": "preprint", "id": "arxiv:2601.17722", "issued": { "date-parts": [ [ 2026, 1, 25 ] ] }, "note": "First published: 2026-01-25; Recorded arXiv version: v1; Recorded version updated: 2026-01-25T06:58:15Z.", "title": "EntWorld: A Holistic Environment and Benchmark for Verifiable Enterprise GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.18092v1", "archive": "arXiv", "archive_location": "2601.18092", "author": [ { "literal": "Nan Chen" }, { "literal": "Jing Lu" }, { "literal": "Zilong Wang" }, { "literal": "Luna K. Qiu" }, { "literal": "Siming Chen" }, { "literal": "Yuqing Yang" } ], "citation-key": "arxiv_2601_18092", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ae1716f673e47cd0e61b5f2225ca4871dfd8522e35037fc8d226f9c7cd0e3e75", "submitted_at": "2026-01-26T03:02:26Z", "updated_at": "2026-01-26T03:02:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.18092v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.18092", "catalog_id": "arxiv:2601.18092", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-01-26", "identifiers": { "arxiv": "2601.18092" }, "record_updated_at": "2026-01-26" }, "genre": "preprint", "id": "arxiv:2601.18092", "issued": { "date-parts": [ [ 2026, 1, 26 ] ] }, "note": "First published: 2026-01-26; Recorded arXiv version: v1; Recorded version updated: 2026-01-26T03:02:26Z.", "title": "From Struggle to Success: Context-Aware Guidance for Screen Reader Users in Computer Use", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.18197v2", "archive": "arXiv", "archive_location": "2601.18197", "author": [ { "literal": "Shaokang Wang" }, { "literal": "Pei Fu" }, { "literal": "Ruoceng Zhang" }, { "literal": "Shaojie Zhang" }, { "literal": "Xiuwen Xi" }, { "literal": "Jiahui Yang" }, { "literal": "Bin Qin" }, { "literal": "Ying Huang" }, { "literal": "Zhenbo Luo" }, { "literal": "Jian Luan" } ], "citation-key": "arxiv_2601_18197", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ccb46b86b0c0b854ed2f1e17f9e12bf031c5ed1349f3d3c44d3276b2a3a195ee", "submitted_at": "2026-01-26T06:29:41Z", "updated_at": "2026-06-26T07:19:55Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.18197v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.18197", "catalog_id": "arxiv:2601.18197", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-26", "identifiers": { "arxiv": "2601.18197" }, "record_updated_at": "2026-06-26" }, "genre": "preprint", "id": "arxiv:2601.18197", "issued": { "date-parts": [ [ 2026, 1, 26 ] ] }, "note": "First published: 2026-01-26; Recorded arXiv version: v2; Recorded version updated: 2026-06-26T07:19:55Z.", "title": "GAIA: A Data Flywheel System for Training GUI Test-Time Scaling Critic Models", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.18305v2", "archive": "arXiv", "archive_location": "2601.18305", "author": [ { "literal": "Xuan Wang" }, { "literal": "Siyuan Su" }, { "literal": "Quantong Fu" }, { "literal": "Yongxiang Hu" }, { "literal": "Yangfan Zhou" } ], "citation-key": "arxiv_2601_18305", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cb90eed33e989d38cd4058d54ba7d073f1b8e5c1453a9106dfe5e533da7b08a1", "submitted_at": "2026-01-26T09:35:10Z", "updated_at": "2026-08-20T07:50:44Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.18305v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.18305", "catalog_id": "arxiv:2601.18305", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-26", "identifiers": { "arxiv": "2601.18305" }, "record_updated_at": "2026-08-20" }, "genre": "preprint", "id": "arxiv:2601.18305", "issued": { "date-parts": [ [ 2026, 1, 26 ] ] }, "note": "First published: 2026-01-26; Recorded arXiv version: v2; Recorded version updated: 2026-08-20T07:50:44Z.", "title": "SwipeGen: Bridging the Execution Gap in GUI Agents via Human-like Swipe Synthesis", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.18842v3", "archive": "arXiv", "archive_location": "2601.18842", "author": [ { "literal": "Yanxi Wang" }, { "literal": "Zhiling Zhang" }, { "literal": "Wenbo Zhou" }, { "literal": "Weiming Zhang" }, { "literal": "Jie Zhang" }, { "literal": "Qiannan Zhu" }, { "literal": "Yu Shi" }, { "literal": "Shuxin Zheng" }, { "literal": "Jiyan He" } ], "citation-key": "arxiv_2601_18842", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1ca912ad67a681e8e908a5a403e982f39b5080bee4df376eb0b598468f3bb504", "submitted_at": "2026-01-26T11:33:40Z", "updated_at": "2026-05-13T07:11:36Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2601.18842v3" }, "canonical_source_url": "https://arxiv.org/abs/2601.18842", "catalog_id": "arxiv:2601.18842", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-26", "identifiers": { "arxiv": "2601.18842" }, "record_updated_at": "2026-05-13" }, "genre": "preprint", "id": "arxiv:2601.18842", "issued": { "date-parts": [ [ 2026, 1, 26 ] ] }, "note": "First published: 2026-01-26; Recorded arXiv version: v3; Recorded version updated: 2026-05-13T07:11:36Z.", "title": "GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2601.19199v2", "archive": "arXiv", "archive_location": "2601.19199", "author": [ { "literal": "Libo Sun" }, { "literal": "Jiwen Zhang" }, { "literal": "Siyuan Wang" }, { "literal": "Zhongyu Wei" } ], "citation-key": "arxiv_2601_19199", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1930b4efca7ee38fc6cd1ceb283c14ed9b1388e7b758b951ba9469ed3067c3e0", "submitted_at": "2026-01-27T04:58:48Z", "updated_at": "2026-01-31T15:17:38Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.19199v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.19199", "catalog_id": "arxiv:2601.19199", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-27", "identifiers": { "arxiv": "2601.19199" }, "record_updated_at": "2026-01-31" }, "genre": "preprint", "id": "arxiv:2601.19199", "issued": { "date-parts": [ [ 2026, 1, 27 ] ] }, "note": "First published: 2026-01-27; Recorded arXiv version: v2; Recorded version updated: 2026-01-31T15:17:38Z.", "title": "MAGNET: Towards Adaptive GUI Agents with Memory-Driven Knowledge Evolution", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.20335v2", "archive": "arXiv", "archive_location": "2601.20335", "author": [ { "literal": "Qinzhuo Wu" }, { "literal": "Zhizhuo Yang" }, { "literal": "Hanhao Li" }, { "literal": "Pengzhi Gao" }, { "literal": "Wei Liu" }, { "literal": "Jian Luan" } ], "citation-key": "arxiv_2601_20335", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1cbf16dd3c790996758e00c085a9c2bccabf5b24f945bbd8440cc1f953e38568", "submitted_at": "2026-01-28T07:49:48Z", "updated_at": "2026-01-29T02:32:46Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.20335v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.20335", "catalog_id": "arxiv:2601.20335", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-28", "identifiers": { "arxiv": "2601.20335" }, "record_updated_at": "2026-01-29" }, "genre": "preprint", "id": "arxiv:2601.20335", "issued": { "date-parts": [ [ 2026, 1, 28 ] ] }, "note": "First published: 2026-01-28; Recorded arXiv version: v2; Recorded version updated: 2026-01-29T02:32:46Z.", "title": "MobileBench-OL: A Comprehensive Chinese Benchmark for Evaluating Mobile GUI Agents in Real-World Environment", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.20380v1", "archive": "arXiv", "archive_location": "2601.20380", "author": [ { "literal": "Le Zhang" }, { "literal": "Yixiong Xiao" }, { "literal": "Xinjiang Lu" }, { "literal": "Jingjia Cao" }, { "literal": "Yusai Zhao" }, { "literal": "Jingbo Zhou" }, { "literal": "Lang An" }, { "literal": "Zikan Feng" }, { "literal": "Wanxiang Sha" }, { "literal": "Yu Shi" }, { "literal": "Congxi Xiao" }, { "literal": "Jian Xiong" }, { "literal": "Yankai Zhang" }, { "literal": "Hua Wu" }, { "literal": "Haifeng Wang" } ], "citation-key": "arxiv_2601_20380", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ad764cf86a2ecab81f3f730d6d26ea582544a355a90a8be55be220c98aabd7bf", "submitted_at": "2026-01-28T08:45:17Z", "updated_at": "2026-01-28T08:45:17Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.20380v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.20380", "catalog_id": "arxiv:2601.20380", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-28", "identifiers": { "arxiv": "2601.20380" }, "record_updated_at": "2026-01-28" }, "genre": "preprint", "id": "arxiv:2601.20380", "issued": { "date-parts": [ [ 2026, 1, 28 ] ] }, "note": "First published: 2026-01-28; Recorded arXiv version: v1; Recorded version updated: 2026-01-28T08:45:17Z.", "title": "OmegaUse: Building a General-Purpose GUI Agent for Autonomous Task Execution", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.20650v3", "archive": "arXiv", "archive_location": "2601.20650", "author": [ { "literal": "Jing Wu" }, { "literal": "Wenjie Ai" }, { "literal": "Daphne Barretto" }, { "literal": "Yiye Chen" }, { "literal": "Qingyu Chen" }, { "literal": "Yuhang He" }, { "literal": "Pranit Chawla" }, { "literal": "Nicholas Gydé" }, { "literal": "Yanan Jian" }, { "literal": "Vibhav Vineet" } ], "citation-key": "arxiv_2601_20650", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0139f7a1402d8dbfdecf65cc53c032495b2084f24cc38ff43841642c9102f29c", "submitted_at": "2026-01-28T14:35:23Z", "updated_at": "2026-08-26T22:31:44Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2601.20650v3" }, "canonical_source_url": "https://arxiv.org/abs/2601.20650", "catalog_id": "arxiv:2601.20650", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-28", "identifiers": { "arxiv": "2601.20650" }, "record_updated_at": "2026-08-26" }, "genre": "preprint", "id": "arxiv:2601.20650", "issued": { "date-parts": [ [ 2026, 1, 28 ] ] }, "note": "First published: 2026-01-28; Recorded arXiv version: v3; Recorded version updated: 2026-08-26T22:31:44Z.", "title": "OS-Marathon: Benchmarking Computer-Use Agents on Vast-Horizon, Repetitive Tasks", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2601.20732v4", "archive": "arXiv", "archive_location": "2601.20732", "author": [ { "literal": "Ziwei Liu" }, { "literal": "Borui Kang" }, { "literal": "Hangjie Yuan" }, { "literal": "Zixiang Zhao" }, { "literal": "Wei Li" }, { "literal": "Yifan Zhu" }, { "literal": "Tao Feng" } ], "citation-key": "arxiv_2601_20732", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1083cf063ccf1781f2cdfe5e266b75b52079fd12ad16d05755577b880c010aec", "submitted_at": "2026-01-28T16:06:31Z", "updated_at": "2026-03-25T10:25:50Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2601.20732v4" }, "canonical_source_url": "https://arxiv.org/abs/2601.20732", "catalog_id": "arxiv:2601.20732", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-28", "identifiers": { "arxiv": "2601.20732" }, "record_updated_at": "2026-03-25" }, "genre": "preprint", "id": "arxiv:2601.20732", "issued": { "date-parts": [ [ 2026, 1, 28 ] ] }, "note": "First published: 2026-01-28; Recorded arXiv version: v4; Recorded version updated: 2026-03-25T10:25:50Z.", "title": "Continual GUI Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2601.21253v1", "archive": "arXiv", "archive_location": "2601.21253", "author": [ { "literal": "Wei Minn" }, { "literal": "Biniam Fisseha Demissie" }, { "literal": "Yan Naing Tun" }, { "literal": "Jiakun Liu" }, { "literal": "Mariano Ceccato" }, { "literal": "Lwin Khin Shar" }, { "literal": "David Lo" } ], "citation-key": "arxiv_2601_21253", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ac3f148047f7e63a33780fed3fd1ca461bead2128914c46f353efab4675df444", "submitted_at": "2026-01-29T04:21:11Z", "updated_at": "2026-01-29T04:21:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.21253v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.21253", "catalog_id": "arxiv:2601.21253", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-29", "identifiers": { "arxiv": "2601.21253" }, "record_updated_at": "2026-01-29" }, "genre": "preprint", "id": "arxiv:2601.21253", "issued": { "date-parts": [ [ 2026, 1, 29 ] ] }, "note": "First published: 2026-01-29; Recorded arXiv version: v1; Recorded version updated: 2026-01-29T04:21:11Z.", "title": "CovAgent: Overcoming the 30% Curse of Mobile Application Coverage with Agentic AI and Dynamic Instrumentation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2601.21872v2", "archive": "arXiv", "archive_location": "2601.21872", "author": [ { "literal": "Yao Zhang" }, { "literal": "Shijie Tang" }, { "literal": "Zeyu Li" }, { "literal": "Zhen Han" }, { "literal": "Volker Tresp" } ], "citation-key": "arxiv_2601_21872", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "68ae67856eb871d9a8f259dfb180bde02d29ee97fd7abf0e60719fc2c7abd7ab", "submitted_at": "2026-01-29T15:39:50Z", "updated_at": "2026-04-09T17:48:11Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.21872v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.21872", "catalog_id": "arxiv:2601.21872", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-29", "identifiers": { "arxiv": "2601.21872" }, "record_updated_at": "2026-04-09" }, "genre": "preprint", "id": "arxiv:2601.21872", "issued": { "date-parts": [ [ 2026, 1, 29 ] ] }, "note": "First published: 2026-01-29; Recorded arXiv version: v2; Recorded version updated: 2026-04-09T17:48:11Z.", "title": "WebArbiter: A Principle-Guided Reasoning Process Reward Model for Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.21961v2", "archive": "arXiv", "archive_location": "2601.21961", "author": [ { "literal": "Kuai Yu" }, { "literal": "Naicheng Yu" }, { "literal": "Han Wang" }, { "literal": "Rui Yang" }, { "literal": "Huan Zhang" } ], "citation-key": "arxiv_2601_21961", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1380697158d3a91334427cf067f3f91a751bd89d31103edecfa68171b28b8d66", "submitted_at": "2026-01-29T16:40:15Z", "updated_at": "2026-02-27T16:47:40Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2601.21961v2" }, "canonical_source_url": "https://arxiv.org/abs/2601.21961", "catalog_id": "arxiv:2601.21961", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-29", "identifiers": { "arxiv": "2601.21961" }, "record_updated_at": "2026-02-27" }, "genre": "preprint", "id": "arxiv:2601.21961", "issued": { "date-parts": [ [ 2026, 1, 29 ] ] }, "note": "First published: 2026-01-29; Recorded arXiv version: v2; Recorded version updated: 2026-02-27T16:47:40Z.", "title": "How do Visual Attributes Influence Web Agents? A Comprehensive Evaluation of User Interface Design Factors", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2601.22781v1", "archive": "arXiv", "archive_location": "2601.22781", "author": [ { "literal": "Linjia Kang" }, { "literal": "Zhimin Wang" }, { "literal": "Yongkang Zhang" }, { "literal": "Duo Wu" }, { "literal": "Jinghe Wang" }, { "literal": "Ming Ma" }, { "literal": "Haopeng Yan" }, { "literal": "Zhi Wang" } ], "citation-key": "arxiv_2601_22781", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0d66365682d0ea4cee744c6c59d3a585c9272a31da5f8ff7c27983d719f070a7", "submitted_at": "2026-01-30T10:03:20Z", "updated_at": "2026-01-30T10:03:20Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2601.22781v1" }, "canonical_source_url": "https://arxiv.org/abs/2601.22781", "catalog_id": "arxiv:2601.22781", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-30", "identifiers": { "arxiv": "2601.22781" }, "record_updated_at": "2026-01-30" }, "genre": "preprint", "id": "arxiv:2601.22781", "issued": { "date-parts": [ [ 2026, 1, 30 ] ] }, "note": "First published: 2026-01-30; Recorded arXiv version: v1; Recorded version updated: 2026-01-30T10:03:20Z.", "title": "Learning with Challenges: Adaptive Difficulty-Aware Data Generation for Mobile GUI Agent Training", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.00575v2", "archive": "arXiv", "archive_location": "2602.00575", "author": [ { "literal": "Chaoqun Cui" }, { "literal": "Jing Huang" }, { "literal": "Shijing Wang" }, { "literal": "Liming Zheng" }, { "literal": "Qingchao Kong" }, { "literal": "Zhixiong Zeng" } ], "citation-key": "arxiv_2602_00575", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "86225f2e52e1394e2a0a56b1e82327d756bff95739e01ce7cc5898af49944aba", "submitted_at": "2026-01-31T07:36:54Z", "updated_at": "2026-07-25T06:54:35Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.00575v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.00575", "catalog_id": "arxiv:2602.00575", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-01-31", "identifiers": { "arxiv": "2602.00575" }, "record_updated_at": "2026-07-25" }, "genre": "preprint", "id": "arxiv:2602.00575", "issued": { "date-parts": [ [ 2026, 1, 31 ] ] }, "note": "First published: 2026-01-31; Recorded arXiv version: v2; Recorded version updated: 2026-07-25T06:54:35Z.", "title": "Agentic Reward Modeling: Verifying GUI Agent via Progressive Trajectory-Grounded Interaction", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.01725v1", "archive": "arXiv", "archive_location": "2602.01725", "author": [ { "literal": "Yurun Chen" }, { "literal": "Zeyi Liao" }, { "literal": "Ping Yin" }, { "literal": "Taotao Xie" }, { "literal": "Keting Yin" }, { "literal": "Shengyu Zhang" } ], "citation-key": "arxiv_2602_01725", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d1a6f98c0ab0a221c169796d9680e4b2a7bf7a0775c0493254f47d5e0a9b3c4e", "submitted_at": "2026-02-02T07:04:06Z", "updated_at": "2026-02-02T07:04:06Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.01725v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.01725", "catalog_id": "arxiv:2602.01725", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-02", "identifiers": { "arxiv": "2602.01725" }, "record_updated_at": "2026-02-02" }, "genre": "preprint", "id": "arxiv:2602.01725", "issued": { "date-parts": [ [ 2026, 2, 2 ] ] }, "note": "First published: 2026-02-02; Recorded arXiv version: v1; Recorded version updated: 2026-02-02T07:04:06Z.", "title": "SafePred: A Predictive Guardrail for Computer-Using Agents via World Models", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.02419v2", "archive": "arXiv", "archive_location": "2602.02419", "author": [ { "literal": "Qingni Wang" }, { "literal": "Yue Fan" }, { "literal": "Xin Eric Wang" } ], "citation-key": "arxiv_2602_02419", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c9557fc608a7fd7667746e9b3e06bc699b1796cdd77fe5e975ba1bd5a64690c1", "submitted_at": "2026-02-02T18:22:45Z", "updated_at": "2026-02-03T08:31:29Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.02419v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.02419", "catalog_id": "arxiv:2602.02419", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-02", "identifiers": { "arxiv": "2602.02419" }, "record_updated_at": "2026-02-03" }, "genre": "preprint", "id": "arxiv:2602.02419", "issued": { "date-parts": [ [ 2026, 2, 2 ] ] }, "note": "First published: 2026-02-02; Recorded arXiv version: v2; Recorded version updated: 2026-02-03T08:31:29Z.", "title": "SafeGround: Know When to Trust GUI Grounding Models via Uncertainty Calibration", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.02995v1", "archive": "arXiv", "archive_location": "2602.02995", "author": [ { "literal": "Sizhe Tang" }, { "literal": "Rongqian Chen" }, { "literal": "Tian Lan" } ], "citation-key": "arxiv_2602_02995", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0a01499ce704bb7916a3157ff2d3fc755a1115d9b27c7e529c153bce360a5404", "submitted_at": "2026-02-03T02:07:12Z", "updated_at": "2026-02-03T02:07:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.02995v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.02995", "catalog_id": "arxiv:2602.02995", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-03", "identifiers": { "arxiv": "2602.02995" }, "record_updated_at": "2026-02-03" }, "genre": "preprint", "id": "arxiv:2602.02995", "issued": { "date-parts": [ [ 2026, 2, 3 ] ] }, "note": "First published: 2026-02-03; Recorded arXiv version: v1; Recorded version updated: 2026-02-03T02:07:12Z.", "title": "Agent Alpha: Tree Search Unifying Generation, Exploration and Evaluation for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.03792v1", "archive": "arXiv", "archive_location": "2602.03792", "author": [ { "literal": "Xilong Wang" }, { "literal": "Yinuo Liu" }, { "literal": "Zhun Wang" }, { "literal": "Dawn Song" }, { "literal": "Neil Gong" } ], "citation-key": "arxiv_2602_03792", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "86702018906e64294d7b8daea618c5f152e8b94b604d27d6b164e82aff462564", "submitted_at": "2026-02-03T17:55:04Z", "updated_at": "2026-02-03T17:55:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.03792v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.03792", "catalog_id": "arxiv:2602.03792", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-03", "identifiers": { "arxiv": "2602.03792" }, "record_updated_at": "2026-02-03" }, "genre": "preprint", "id": "arxiv:2602.03792", "issued": { "date-parts": [ [ 2026, 2, 3 ] ] }, "note": "First published: 2026-02-03; Recorded arXiv version: v1; Recorded version updated: 2026-02-03T17:55:04Z.", "title": "WebSentinel: Detecting and Localizing Prompt Injection Attacks for Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.05354v3", "archive": "arXiv", "archive_location": "2602.05354", "author": [ { "literal": "Syed Nazmus Sakib" }, { "literal": "Nafiul Haque" }, { "literal": "Tapodhir Karmakar Taton" }, { "literal": "Shahrear Bin Amin" }, { "literal": "Shifat E. Arman" } ], "citation-key": "arxiv_2602_05354", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "76fbeae086fb9ef21f820b02777a2ceb66ee37f8f1c978c7426589ea4e888e49", "submitted_at": "2026-02-05T06:24:23Z", "updated_at": "2026-09-08T14:40:01Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2602.05354v3" }, "canonical_source_url": "https://arxiv.org/abs/2602.05354", "catalog_id": "arxiv:2602.05354", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-05", "identifiers": { "arxiv": "2602.05354" }, "record_updated_at": "2026-09-08" }, "genre": "preprint", "id": "arxiv:2602.05354", "issued": { "date-parts": [ [ 2026, 2, 5 ] ] }, "note": "First published: 2026-02-05; Recorded arXiv version: v3; Recorded version updated: 2026-09-08T14:40:01Z.", "title": "Do Web Agents Investigate Before They Decide?", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2602.06075v3", "archive": "arXiv", "archive_location": "2602.06075", "author": [ { "literal": "Guangyi Liu" }, { "literal": "Pengxiang Zhao" }, { "literal": "Yaozhen Liang" }, { "literal": "Qinyi Luo" }, { "literal": "Shunye Tang" }, { "literal": "Yuxiang Chai" }, { "literal": "Weifeng Lin" }, { "literal": "Han Xiao" }, { "literal": "WenHao Wang" }, { "literal": "Siheng Chen" }, { "literal": "Zhengxi Lu" }, { "literal": "Gao Wu" }, { "literal": "Hao Wang" }, { "literal": "Liang Liu" }, { "literal": "Yong Liu" } ], "citation-key": "arxiv_2602_06075", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0692d6936a3e3c20a7ad61c69ae770f5669c3783ff837092254f8b3fb170cd0a", "submitted_at": "2026-02-03T17:01:59Z", "updated_at": "2026-08-27T06:48:18Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2602.06075v3" }, "canonical_source_url": "https://arxiv.org/abs/2602.06075", "catalog_id": "arxiv:2602.06075", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-03", "identifiers": { "arxiv": "2602.06075" }, "record_updated_at": "2026-08-27" }, "genre": "preprint", "id": "arxiv:2602.06075", "issued": { "date-parts": [ [ 2026, 2, 3 ] ] }, "note": "First published: 2026-02-03; Recorded arXiv version: v3; Recorded version updated: 2026-08-27T06:48:18Z.", "title": "MemGUI-Bench: Benchmarking Memory of Mobile GUI Agents in Dynamic Environments", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2602.07153v2", "archive": "arXiv", "archive_location": "2602.07153", "author": [ { "literal": "Jinbiao Wei" }, { "literal": "Yilun Zhao" }, { "literal": "Kangqi Ni" }, { "literal": "Arman Cohan" } ], "citation-key": "arxiv_2602_07153", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ce896d4d846bf07fedc68b7f952532c4fefe24dcb49b51556e0d6c2bc648bb91", "submitted_at": "2026-02-06T19:55:26Z", "updated_at": "2026-04-12T06:09:23Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.07153v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.07153", "catalog_id": "arxiv:2602.07153", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-06", "identifiers": { "arxiv": "2602.07153" }, "record_updated_at": "2026-04-12" }, "genre": "preprint", "id": "arxiv:2602.07153", "issued": { "date-parts": [ [ 2026, 2, 6 ] ] }, "note": "First published: 2026-02-06; Recorded arXiv version: v2; Recorded version updated: 2026-04-12T06:09:23Z.", "title": "ANCHOR: Branch-Point Data Generation for GUI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.07283v1", "archive": "arXiv", "archive_location": "2602.07283", "author": [ { "literal": "Ruijia Cheng" }, { "literal": "Jenny T. Liang" }, { "literal": "Eldon Schoop" }, { "literal": "Jeffrey Nichols" } ], "citation-key": "arxiv_2602_07283", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d556ebbfbf4fc5875395b449a92b8ff27a7ed59d41bc062ac60fcf175f9064f8", "submitted_at": "2026-02-07T00:19:08Z", "updated_at": "2026-02-07T00:19:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.07283v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.07283", "catalog_id": "arxiv:2602.07283", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-07", "identifiers": { "arxiv": "2602.07283" }, "record_updated_at": "2026-02-07" }, "genre": "preprint", "id": "arxiv:2602.07283", "issued": { "date-parts": [ [ 2026, 2, 7 ] ] }, "note": "First published: 2026-02-07; Recorded arXiv version: v1; Recorded version updated: 2026-02-07T00:19:08Z.", "title": "Mapping the Design Space of User Experience for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.08995v2", "archive": "arXiv", "archive_location": "2602.08995", "author": [ { "literal": "Yuting Ning" }, { "literal": "Jaylen Jones" }, { "literal": "Zhehao Zhang" }, { "literal": "Chentao Ye" }, { "literal": "Weitong Ruan" }, { "literal": "Junyi Li" }, { "literal": "Rahul Gupta" }, { "literal": "Huan Sun" } ], "citation-key": "arxiv_2602_08995", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "853f64c020f47129be01778d5212ade8ef78251959ad634ee04ad0fb9fe2989c", "submitted_at": "2026-02-09T18:41:15Z", "updated_at": "2026-06-25T04:36:12Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.08995v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.08995", "catalog_id": "arxiv:2602.08995", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-09", "identifiers": { "arxiv": "2602.08995" }, "record_updated_at": "2026-06-25" }, "genre": "preprint", "id": "arxiv:2602.08995", "issued": { "date-parts": [ [ 2026, 2, 9 ] ] }, "note": "First published: 2026-02-09; Recorded arXiv version: v2; Recorded version updated: 2026-06-25T04:36:12Z.", "title": "When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.09007v3", "archive": "arXiv", "archive_location": "2602.09007", "author": [ { "literal": "Haodong Li" }, { "literal": "Jingwei Wu" }, { "literal": "Quan Sun" }, { "literal": "Guopeng Li" }, { "literal": "Juanxi Tian" }, { "literal": "Ziyu Guo" }, { "literal": "Donghao Zhou" }, { "literal": "Tianfei Ren" }, { "literal": "Huanyu Zhang" }, { "literal": "Yanlin Lai" }, { "literal": "Ruichuan An" }, { "literal": "Hongbo Peng" }, { "literal": "Yuhong Dai" }, { "literal": "Jia Wang" }, { "literal": "Ziyang Meng" }, { "literal": "Zheng Ge" }, { "literal": "Xiangyu Zhang" }, { "literal": "Daxin Jiang" }, { "literal": "Pheng-Ann Heng" } ], "citation-key": "arxiv_2602_09007", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "43a01e0eb848a57a84767836d2c3c982bbf6094d22a3e70652dd42af52760bc5", "submitted_at": "2026-02-09T18:52:02Z", "updated_at": "2026-09-26T18:28:57Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2602.09007v3" }, "canonical_source_url": "https://arxiv.org/abs/2602.09007", "catalog_id": "arxiv:2602.09007", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-09", "identifiers": { "arxiv": "2602.09007" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2602.09007", "issued": { "date-parts": [ [ 2026, 2, 9 ] ] }, "note": "First published: 2026-02-09; Recorded arXiv version: v3; Recorded version updated: 2026-09-26T18:28:57Z.", "title": "GUI-GenBench: Evaluating Image Generation Models as Interactive GUI Environments", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2602.09012v2", "archive": "arXiv", "archive_location": "2602.09012", "author": [ { "literal": "Jiacheng Liu" }, { "literal": "Yaxin Luo" }, { "literal": "Jiacheng Cui" }, { "literal": "Xinyi Shang" }, { "literal": "Xiaohan Zhao" }, { "literal": "Zhiqiang Shen" } ], "citation-key": "arxiv_2602_09012", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3178684a322fcf07f847473139efdddf0b02213af468f1afbd285549d15bde56", "submitted_at": "2026-02-09T18:55:33Z", "updated_at": "2026-06-21T08:25:39Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.09012v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.09012", "catalog_id": "arxiv:2602.09012", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-09", "identifiers": { "arxiv": "2602.09012" }, "record_updated_at": "2026-06-21" }, "genre": "preprint", "id": "arxiv:2602.09012", "issued": { "date-parts": [ [ 2026, 2, 9 ] ] }, "note": "First published: 2026-02-09; Recorded arXiv version: v2; Recorded version updated: 2026-06-21T08:25:39Z.", "title": "Next-Gen CAPTCHAs: Leveraging the Cognitive Gap for Scalable and Diverse GUI-Agent Defense", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.09310v1", "archive": "arXiv", "archive_location": "2602.09310", "author": [ { "literal": "Ananya Gubbi Mohanbabu" }, { "literal": "Rosiana Natalie" }, { "literal": "Brandon Kim" }, { "literal": "Anhong Guo" }, { "literal": "Amy Pavel" } ], "citation-key": "arxiv_2602_09310", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c3606d090749737d4223a9f826400c2b2aebde245c623391f2a8c6702b2468f8", "submitted_at": "2026-02-10T01:00:55Z", "updated_at": "2026-02-10T01:00:55Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.09310v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.09310", "catalog_id": "arxiv:2602.09310", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-10", "identifiers": { "arxiv": "2602.09310" }, "record_updated_at": "2026-02-10" }, "genre": "preprint", "id": "arxiv:2602.09310", "issued": { "date-parts": [ [ 2026, 2, 10 ] ] }, "note": "First published: 2026-02-10; Recorded arXiv version: v1; Recorded version updated: 2026-02-10T01:00:55Z.", "title": "A11y-CUA Dataset: Characterizing the Accessibility Gap in Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.10814v1", "archive": "arXiv", "archive_location": "2602.10814", "author": [ { "literal": "Xingyi Zhang" }, { "literal": "Yulei Ye" }, { "literal": "Kaifeng Huang" }, { "literal": "Wenhao Li" }, { "literal": "Xiangfeng Wang" } ], "citation-key": "arxiv_2602_10814", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "86a47f084b23cc70e754808f6404bed0bbd500f78fcac4ccc104f95052a49868", "submitted_at": "2026-02-11T12:54:53Z", "updated_at": "2026-02-11T12:54:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.10814v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.10814", "catalog_id": "arxiv:2602.10814", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-11", "identifiers": { "arxiv": "2602.10814" }, "record_updated_at": "2026-02-11" }, "genre": "preprint", "id": "arxiv:2602.10814", "issued": { "date-parts": [ [ 2026, 2, 11 ] ] }, "note": "First published: 2026-02-11; Recorded arXiv version: v1; Recorded version updated: 2026-02-11T12:54:53Z.", "title": "See, Plan, Snap: Evaluating Multimodal GUI Agents in Scratch", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.11724v3", "archive": "arXiv", "archive_location": "2602.11724", "author": [ { "literal": "Xiwen Teoh" }, { "literal": "Yun Lin" }, { "literal": "Duc-Minh Nguyen" }, { "literal": "Ruofei Ren" }, { "literal": "Wenjie Zhang" }, { "literal": "Jin Song Dong" } ], "citation-key": "arxiv_2602_11724", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2076fc8b0b7b37bdf43fe2e0b0cc5fa4c3a8cb7b44a41fdf80f59a8c05c7e1b3", "submitted_at": "2026-02-12T08:51:07Z", "updated_at": "2026-04-23T15:00:58Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2602.11724v3" }, "canonical_source_url": "https://arxiv.org/abs/2602.11724", "catalog_id": "arxiv:2602.11724", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-12", "identifiers": { "arxiv": "2602.11724" }, "record_updated_at": "2026-04-23" }, "genre": "preprint", "id": "arxiv:2602.11724", "issued": { "date-parts": [ [ 2026, 2, 12 ] ] }, "note": "First published: 2026-02-12; Recorded arXiv version: v3; Recorded version updated: 2026-04-23T15:00:58Z.", "title": "WebTestPilot: Agentic End-to-End Web Testing against Natural Language Specification by Inferring Oracles with Symbolized GUI Elements", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2602.11750v1", "archive": "arXiv", "archive_location": "2602.11750", "author": [ { "literal": "Jiazheng Sun" }, { "literal": "Mingxuan Li" }, { "literal": "Yingying Zhang" }, { "literal": "Jiayang Niu" }, { "literal": "Yachen Wu" }, { "literal": "Ruihan Jin" }, { "literal": "Shuyu Lei" }, { "literal": "Pengrongrui Tan" }, { "literal": "Zongyu Zhang" }, { "literal": "Ruoyi Wang" }, { "literal": "Jiachen Yang" }, { "literal": "Boyu Yang" }, { "literal": "Jiacheng Liu" }, { "literal": "Xin Peng" } ], "citation-key": "arxiv_2602_11750", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2bbba09ca845f4fb772516d26d59fc3d99f8e87ba3657bb29c3c062b50925ec5", "submitted_at": "2026-02-12T09:25:15Z", "updated_at": "2026-02-12T09:25:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.11750v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.11750", "catalog_id": "arxiv:2602.11750", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-12", "identifiers": { "arxiv": "2602.11750" }, "record_updated_at": "2026-02-12" }, "genre": "preprint", "id": "arxiv:2602.11750", "issued": { "date-parts": [ [ 2026, 2, 12 ] ] }, "note": "First published: 2026-02-12; Recorded arXiv version: v1; Recorded version updated: 2026-02-12T09:25:15Z.", "title": "AmbiBench: Benchmarking Mobile GUI Agents Beyond One-Shot Instructions in the Wild", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.12276v2", "archive": "arXiv", "archive_location": "2602.12276", "author": [ { "literal": "Nicholas Lee" }, { "literal": "Lutfi Eren Erdogan" }, { "literal": "Chris Joseph John" }, { "literal": "Surya Krishnapillai" }, { "literal": "Michael W. Mahoney" }, { "literal": "Kurt Keutzer" }, { "literal": "Amir Gholami" } ], "citation-key": "arxiv_2602_12276", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c64cd65fc7fec601658baf7836bdbe9c0e494bde2b0a4b66ec9d3090ee106b1f", "submitted_at": "2026-02-12T18:58:30Z", "updated_at": "2026-08-14T21:25:10Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.12276v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.12276", "catalog_id": "arxiv:2602.12276", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-12", "identifiers": { "arxiv": "2602.12276" }, "record_updated_at": "2026-08-14" }, "genre": "preprint", "id": "arxiv:2602.12276", "issued": { "date-parts": [ [ 2026, 2, 12 ] ] }, "note": "First published: 2026-02-12; Recorded arXiv version: v2; Recorded version updated: 2026-08-14T21:25:10Z.", "title": "Agentic Test-Time Scaling for WebAgents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.12544v1", "archive": "arXiv", "archive_location": "2602.12544", "author": [ { "literal": "Lajanugen Logeswaran" }, { "literal": "Jaekyeom Kim" }, { "literal": "Sungryull Sohn" }, { "literal": "Creighton Glasscock" }, { "literal": "Honglak Lee" } ], "citation-key": "arxiv_2602_12544", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "555285a1ab51e02661067bdf8bcc3fa7f6ee0fdd63f52f786a03eb2483bd1a31", "submitted_at": "2026-02-13T02:52:18Z", "updated_at": "2026-02-13T02:52:18Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.12544v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.12544", "catalog_id": "arxiv:2602.12544", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-13", "identifiers": { "arxiv": "2602.12544" }, "record_updated_at": "2026-02-13" }, "genre": "preprint", "id": "arxiv:2602.12544", "issued": { "date-parts": [ [ 2026, 2, 13 ] ] }, "note": "First published: 2026-02-13; Recorded arXiv version: v1; Recorded version updated: 2026-02-13T02:52:18Z.", "title": "Scaling Web Agent Training through Automatic Data Generation and Fine-grained Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.13516v1", "archive": "arXiv", "archive_location": "2602.13516", "author": [ { "literal": "Jaechul Roh" }, { "literal": "Eugene Bagdasarian" }, { "literal": "Hamed Haddadi" }, { "literal": "Ali Shahin Shamsabadi" } ], "citation-key": "arxiv_2602_13516", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fb048b4e5d1d244d689957f30cdcb93cf4dbb750b24ae527b62ecfb04b554eb4", "submitted_at": "2026-02-13T23:02:50Z", "updated_at": "2026-02-13T23:02:50Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.13516v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.13516", "catalog_id": "arxiv:2602.13516", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-13", "identifiers": { "arxiv": "2602.13516" }, "record_updated_at": "2026-02-13" }, "genre": "preprint", "id": "arxiv:2602.13516", "issued": { "date-parts": [ [ 2026, 2, 13 ] ] }, "note": "First published: 2026-02-13; Recorded arXiv version: v1; Recorded version updated: 2026-02-13T23:02:50Z.", "title": "SPILLage: Agentic Oversharing on the Web", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.14093v1", "archive": "arXiv", "archive_location": "2602.14093", "author": [ { "literal": "Yuan Cao" }, { "literal": "Dezhi Ran" }, { "literal": "Mengzhou Wu" }, { "literal": "Yuzhe Guo" }, { "literal": "Xin Chen" }, { "literal": "Ang Li" }, { "literal": "Gang Cao" }, { "literal": "Gong Zhi" }, { "literal": "Hao Yu" }, { "literal": "Linyi Li" }, { "literal": "Wei Yang" }, { "literal": "Tao Xie" } ], "citation-key": "arxiv_2602_14093", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "29b4c8a5bda1df237fad33ce483018b3476684f904841ed0598ec4286234b858", "submitted_at": "2026-02-15T10:58:01Z", "updated_at": "2026-02-15T10:58:01Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.14093v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.14093", "catalog_id": "arxiv:2602.14093", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-15", "identifiers": { "arxiv": "2602.14093" }, "record_updated_at": "2026-02-15" }, "genre": "preprint", "id": "arxiv:2602.14093", "issued": { "date-parts": [ [ 2026, 2, 15 ] ] }, "note": "First published: 2026-02-15; Recorded arXiv version: v1; Recorded version updated: 2026-02-15T10:58:01Z.", "title": "GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.14276v2", "archive": "arXiv", "archive_location": "2602.14276", "author": [ { "literal": "A. Said Gurbuz" }, { "literal": "Sunghwan Hong" }, { "literal": "Ahmed Nassar" }, { "literal": "Marc Pollefeys" }, { "literal": "Peter Staar" } ], "citation-key": "arxiv_2602_14276", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d7c6e23f704562c4ae4db9e0b4f8f2a8dde1318efc08876ee57f5967b178257e", "submitted_at": "2026-02-15T19:00:02Z", "updated_at": "2026-05-01T17:32:06Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.14276v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.14276", "catalog_id": "arxiv:2602.14276", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-15", "identifiers": { "arxiv": "2602.14276" }, "record_updated_at": "2026-05-01" }, "genre": "preprint", "id": "arxiv:2602.14276", "issued": { "date-parts": [ [ 2026, 2, 15 ] ] }, "note": "First published: 2026-02-15; Recorded arXiv version: v2; Recorded version updated: 2026-05-01T17:32:06Z.", "title": "ScreenParse: Moving Beyond Sparse Grounding with Complete Screen Parsing Supervision", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.14296v1", "archive": "arXiv", "archive_location": "2602.14296", "author": [ { "literal": "Yifan Wu" }, { "literal": "Yiran Peng" }, { "literal": "Yiyu Chen" }, { "literal": "Jianhao Ruan" }, { "literal": "Zijie Zhuang" }, { "literal": "Cheng Yang" }, { "literal": "Jiayi Zhang" }, { "literal": "Man Chen" }, { "literal": "Yenchi Tseng" }, { "literal": "Zhaoyang Yu" }, { "literal": "Liang Chen" }, { "literal": "Yuyao Zhai" }, { "literal": "Bang Liu" }, { "literal": "Chenglin Wu" }, { "literal": "Yuyu Luo" } ], "citation-key": "arxiv_2602_14296", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3773ec51f5316b9006f93aeafd7b654bf0e7418253c0af16ef7b7c00fddea102", "submitted_at": "2026-02-15T20:03:19Z", "updated_at": "2026-02-15T20:03:19Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.14296v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.14296", "catalog_id": "arxiv:2602.14296", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-15", "identifiers": { "arxiv": "2602.14296" }, "record_updated_at": "2026-02-15" }, "genre": "preprint", "id": "arxiv:2602.14296", "issued": { "date-parts": [ [ 2026, 2, 15 ] ] }, "note": "First published: 2026-02-15; Recorded arXiv version: v1; Recorded version updated: 2026-02-15T20:03:19Z.", "title": "AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State Machines", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.14721v1", "archive": "arXiv", "archive_location": "2602.14721", "author": [ { "literal": "Zikai Xiao" }, { "literal": "Jianhong Tu" }, { "literal": "Chuhang Zou" }, { "literal": "Yuxin Zuo" }, { "literal": "Zhi Li" }, { "literal": "Peng Wang" }, { "literal": "Bowen Yu" }, { "literal": "Fei Huang" }, { "literal": "Junyang Lin" }, { "literal": "Zuozhu Liu" } ], "citation-key": "arxiv_2602_14721", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2dca0e4219a788205bbb060911d0f004d1db4cf2ff8a4962c04fd5777ea194d7", "submitted_at": "2026-02-16T13:06:49Z", "updated_at": "2026-02-16T13:06:49Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2602.14721v1" }, "canonical_source_url": "https://arxiv.org/abs/2602.14721", "catalog_id": "arxiv:2602.14721", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-16", "identifiers": { "arxiv": "2602.14721" }, "record_updated_at": "2026-02-16" }, "genre": "preprint", "id": "arxiv:2602.14721", "issued": { "date-parts": [ [ 2026, 2, 16 ] ] }, "note": "First published: 2026-02-16; Recorded arXiv version: v1; Recorded version updated: 2026-02-16T13:06:49Z.", "title": "WebWorld: A Large-Scale World Model for Web Agent Training", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2602.16313v2", "archive": "arXiv", "archive_location": "2602.16313", "author": [ { "literal": "Zexue He" }, { "literal": "Yu Wang" }, { "literal": "Churan Zhi" }, { "literal": "Yuanzhe Hu" }, { "literal": "Tzu-Ping Chen" }, { "literal": "Lang Yin" }, { "literal": "Ze Chen" }, { "literal": "Tong Arthur Wu" }, { "literal": "Siru Ouyang" }, { "literal": "Zihan Wang" }, { "literal": "Jiaxin Pei" }, { "literal": "Julian McAuley" }, { "literal": "Yejin Choi" }, { "literal": "Alex Pentland" } ], "citation-key": "arxiv_2602_16313", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2317a94d54b4458a1b7c5e92754aeaefcbf270aabfa0bd8ea333c673cd4596b3", "submitted_at": "2026-02-18T09:49:14Z", "updated_at": "2026-09-17T22:55:03Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2602.16313v2" }, "canonical_source_url": "https://arxiv.org/abs/2602.16313", "catalog_id": "arxiv:2602.16313", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-18", "identifiers": { "arxiv": "2602.16313" }, "record_updated_at": "2026-09-17" }, "genre": "preprint", "id": "arxiv:2602.16313", "issued": { "date-parts": [ [ 2026, 2, 18 ] ] }, "note": "First published: 2026-02-18; Recorded arXiv version: v2; Recorded version updated: 2026-09-17T22:55:03Z.", "title": "MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2602.17003v3", "archive": "arXiv", "archive_location": "2602.17003", "author": [ { "literal": "Serin Kim" }, { "literal": "Sangam Lee" }, { "literal": "Dongha Lee" } ], "citation-key": "arxiv_2602_17003", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "28a23bf22920f8c33317e9d16f59d7085f188e79157301ff573abe773cf562ce", "submitted_at": "2026-02-19T01:54:26Z", "updated_at": "2026-05-27T12:15:49Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2602.17003v3" }, "canonical_source_url": "https://arxiv.org/abs/2602.17003", "catalog_id": "arxiv:2602.17003", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-19", "identifiers": { "arxiv": "2602.17003" }, "record_updated_at": "2026-05-27" }, "genre": "preprint", "id": "arxiv:2602.17003", "issued": { "date-parts": [ [ 2026, 2, 19 ] ] }, "note": "First published: 2026-02-19; Recorded arXiv version: v3; Recorded version updated: 2026-05-27T12:15:49Z.", "title": "Persona2Web: Benchmarking Personalized Web Agents for Contextual Reasoning with User History", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2603.02951v2", "archive": "arXiv", "archive_location": "2603.02951", "author": [ { "literal": "Zhenquan Yao" }, { "literal": "Zitong Huang" }, { "literal": "Yihan Zeng" }, { "literal": "Jianhua Han" }, { "literal": "Hang Xu" }, { "literal": "Chun-Mei Feng" }, { "literal": "Jianwei Ma" }, { "literal": "Wangmeng Zuo" } ], "citation-key": "arxiv_2603_02951", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "75f8eb625e969c2b180eeb3e90aa4e513ab5a1dc15657d3849c68b69bc5edf7c", "submitted_at": "2026-03-03T13:02:20Z", "updated_at": "2026-03-07T10:11:08Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2603.02951v2" }, "canonical_source_url": "https://arxiv.org/abs/2603.02951", "catalog_id": "arxiv:2603.02951", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-03", "identifiers": { "arxiv": "2603.02951" }, "record_updated_at": "2026-03-07" }, "genre": "preprint", "id": "arxiv:2603.02951", "issued": { "date-parts": [ [ 2026, 3, 3 ] ] }, "note": "First published: 2026-03-03; Recorded arXiv version: v2; Recorded version updated: 2026-03-07T10:11:08Z.", "title": "CGL: Advancing Continual GUI Learning via Reinforcement Fine-Tuning", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2603.04364v2", "archive": "arXiv", "archive_location": "2603.04364", "author": [ { "literal": "Haoyu Liu" }, { "literal": "Dingcheng Li" }, { "literal": "Lukas Rutishauser" }, { "literal": "Zeyu Zheng" } ], "citation-key": "arxiv_2603_04364", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "971770e61bf0af91ad1f17b197f09fc166b26bfb4953d3f2c74d457a87e25f3a", "submitted_at": "2026-03-04T18:29:54Z", "updated_at": "2026-10-05T23:01:49Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2603.04364v2" }, "canonical_source_url": "https://arxiv.org/abs/2603.04364", "catalog_id": "arxiv:2603.04364", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-04", "identifiers": { "arxiv": "2603.04364" }, "record_updated_at": "2026-10-05" }, "genre": "preprint", "id": "arxiv:2603.04364", "issued": { "date-parts": [ [ 2026, 3, 4 ] ] }, "note": "First published: 2026-03-04; Recorded arXiv version: v2; Recorded version updated: 2026-10-05T23:01:49Z.", "title": "Dual-Modality Multi-Stage Adversarial Safety Training: Robustifying Multimodal Web Agents Against Cross-Modal Attacks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2603.04949v4", "archive": "arXiv", "archive_location": "2603.04949", "author": [ { "literal": "Md Farhan Ishmam" }, { "literal": "Kenneth Marino" } ], "citation-key": "arxiv_2603_04949", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1901125b21431e41eacb5d5f0fc7a439d9a637c43762d441ece46a90e135051b", "submitted_at": "2026-03-05T08:43:06Z", "updated_at": "2026-10-14T10:16:21Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2603.04949v4" }, "canonical_source_url": "https://arxiv.org/abs/2603.04949", "catalog_id": "arxiv:2603.04949", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-05", "identifiers": { "arxiv": "2603.04949" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2603.04949", "issued": { "date-parts": [ [ 2026, 3, 5 ] ] }, "note": "First published: 2026-03-05; Recorded arXiv version: v4; Recorded version updated: 2026-10-14T10:16:21Z.", "title": "TimeWarp: Evaluating Web Agents by Revisiting the Past", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2603.05295v3", "archive": "arXiv", "archive_location": "2603.05295", "author": [ { "literal": "Sicheng Fan" }, { "literal": "Rui Wan" }, { "literal": "Yifei Leng" }, { "literal": "Gaoning Liang" }, { "literal": "Li Ling" }, { "literal": "Yanyi Shang" }, { "literal": "Dehan Kong" } ], "citation-key": "arxiv_2603_05295", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "db52246d9f5f305046e6dd14f2ef15440c371678e1db0d05282c84a3667c6f10", "submitted_at": "2026-03-05T15:37:34Z", "updated_at": "2026-04-14T08:41:56Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2603.05295v3" }, "canonical_source_url": "https://arxiv.org/abs/2603.05295", "catalog_id": "arxiv:2603.05295", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-05", "identifiers": { "arxiv": "2603.05295" }, "record_updated_at": "2026-04-14" }, "genre": "preprint", "id": "arxiv:2603.05295", "issued": { "date-parts": [ [ 2026, 3, 5 ] ] }, "note": "First published: 2026-03-05; Recorded arXiv version: v3; Recorded version updated: 2026-04-14T08:41:56Z.", "title": "WebChain: A Large-Scale Human-Annotated Dataset of Real-World Web Interaction Traces", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2603.07432v1", "archive": "arXiv", "archive_location": "2603.07432", "author": [ { "literal": "Li Gu" }, { "literal": "Zihuan Jiang" }, { "literal": "Zhixiang Chi" }, { "literal": "Huan Liu" }, { "literal": "Ziqiang Wang" }, { "literal": "Yuanhao Yu" }, { "literal": "Glen Berseth" }, { "literal": "Yang Wang" } ], "citation-key": "arxiv_2603_07432", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "68350312fecc36630ecaeec18a203e692e072d31ff80f442662d202b43ead788", "submitted_at": "2026-03-08T03:08:11Z", "updated_at": "2026-03-08T03:08:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.07432v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.07432", "catalog_id": "arxiv:2603.07432", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-08", "identifiers": { "arxiv": "2603.07432" }, "record_updated_at": "2026-03-08" }, "genre": "preprint", "id": "arxiv:2603.07432", "issued": { "date-parts": [ [ 2026, 3, 8 ] ] }, "note": "First published: 2026-03-08; Recorded arXiv version: v1; Recorded version updated: 2026-03-08T03:08:11Z.", "title": "Generalization in Online Reinforcement Learning for Mobile Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.07978v1", "archive": "arXiv", "archive_location": "2603.07978", "author": [ { "literal": "Jiateng Liu" }, { "literal": "Zhenhailong Wang" }, { "literal": "Rushi Wang" }, { "literal": "Bingxuan Li" }, { "literal": "Jeonghwan Kim" }, { "literal": "Aditi Tiwari" }, { "literal": "Pengfei Yu" }, { "literal": "Denghui Zhang" }, { "literal": "Heng Ji" } ], "citation-key": "arxiv_2603_07978", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "77eb68f31d616b8e542dbb41ada67f28bdc21f29f7da3f04a0077c0fa2d10196", "submitted_at": "2026-03-09T05:27:56Z", "updated_at": "2026-03-09T05:27:56Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.07978v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.07978", "catalog_id": "arxiv:2603.07978", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-09", "identifiers": { "arxiv": "2603.07978" }, "record_updated_at": "2026-03-09" }, "genre": "preprint", "id": "arxiv:2603.07978", "issued": { "date-parts": [ [ 2026, 3, 9 ] ] }, "note": "First published: 2026-03-09; Recorded arXiv version: v1; Recorded version updated: 2026-03-09T05:27:56Z.", "title": "OSExpert: Computer-Use Agents Learning Professional Skills via Exploration", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.08013v1", "archive": "arXiv", "archive_location": "2603.08013", "author": [ { "literal": "Yuxiang Chai" }, { "literal": "Shunye Tang" }, { "literal": "Han Xiao" }, { "literal": "Rui Liu" }, { "literal": "Hongsheng Li" } ], "citation-key": "arxiv_2603_08013", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7334e01b1fbf5cdd02c275c73d2a1b5c2931718d6019fcf6a0ea72b449704d94", "submitted_at": "2026-03-09T06:41:32Z", "updated_at": "2026-03-09T06:41:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.08013v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.08013", "catalog_id": "arxiv:2603.08013", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-09", "identifiers": { "arxiv": "2603.08013" }, "record_updated_at": "2026-03-09" }, "genre": "preprint", "id": "arxiv:2603.08013", "issued": { "date-parts": [ [ 2026, 3, 9 ] ] }, "note": "First published: 2026-03-09; Recorded arXiv version: v1; Recorded version updated: 2026-03-09T06:41:32Z.", "title": "PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.10178v1", "archive": "arXiv", "archive_location": "2603.10178", "author": [ { "literal": "Linxin Song" }, { "literal": "Jieyu Zhang" }, { "literal": "Huanxin Sheng" }, { "literal": "Taiwei Shi" }, { "literal": "Gupta Rahul" }, { "literal": "Yang Liu" }, { "literal": "Ranjay Krishna" }, { "literal": "Jian Kang" }, { "literal": "Jieyu Zhao" } ], "citation-key": "arxiv_2603_10178", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "22b3001540ac8cc75117cdfb4f381462fe7df3fe7700a86e14b9f522803316ac", "submitted_at": "2026-03-10T19:17:22Z", "updated_at": "2026-03-10T19:17:22Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.10178v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.10178", "catalog_id": "arxiv:2603.10178", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-10", "identifiers": { "arxiv": "2603.10178" }, "record_updated_at": "2026-03-10" }, "genre": "preprint", "id": "arxiv:2603.10178", "issued": { "date-parts": [ [ 2026, 3, 10 ] ] }, "note": "First published: 2026-03-10; Recorded arXiv version: v1; Recorded version updated: 2026-03-10T19:17:22Z.", "title": "Video-Based Reward Modeling for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.10268v1", "archive": "arXiv", "archive_location": "2603.10268", "author": [ { "literal": "Syed Yusuf Ahmed" }, { "literal": "Shiwei Feng" }, { "literal": "Chanwoo Bae" }, { "literal": "Calix Barrus Xiangyu Zhang" } ], "citation-key": "arxiv_2603_10268", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8ad18d7fe815f5529fef663130b7ce8841763589e7f4319c0c7556ec2df1da03", "submitted_at": "2026-03-10T22:56:03Z", "updated_at": "2026-03-10T22:56:03Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.10268v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.10268", "catalog_id": "arxiv:2603.10268", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-10", "identifiers": { "arxiv": "2603.10268" }, "record_updated_at": "2026-03-10" }, "genre": "preprint", "id": "arxiv:2603.10268", "issued": { "date-parts": [ [ 2026, 3, 10 ] ] }, "note": "First published: 2026-03-10; Recorded arXiv version: v1; Recorded version updated: 2026-03-10T22:56:03Z.", "title": "SpecOps: A Fully Automated AI Agent Testing Framework in Real-World GUI Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.10577v2", "archive": "arXiv", "archive_location": "2603.10577", "author": [ { "literal": "Marta Sumyk" }, { "literal": "Oleksandr Kosovan" } ], "citation-key": "arxiv_2603_10577", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "048dc7332372585872f5fd3da47a10bd999c7520b2999e45ea6f578b7c856400", "submitted_at": "2026-03-11T09:28:41Z", "updated_at": "2026-03-12T07:30:09Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2603.10577v2" }, "canonical_source_url": "https://arxiv.org/abs/2603.10577", "catalog_id": "arxiv:2603.10577", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-11", "identifiers": { "arxiv": "2603.10577" }, "record_updated_at": "2026-03-12" }, "genre": "preprint", "id": "arxiv:2603.10577", "issued": { "date-parts": [ [ 2026, 3, 11 ] ] }, "note": "First published: 2026-03-11; Recorded arXiv version: v2; Recorded version updated: 2026-03-12T07:30:09Z.", "title": "CUAAudit: Meta-Evaluation of Vision-Language Models as Auditors of Autonomous Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2603.12823v1", "archive": "arXiv", "archive_location": "2603.12823", "author": [ { "literal": "Xunzhuo Liu" }, { "literal": "Bowei He" }, { "literal": "Xue Liu" }, { "literal": "Andy Luo" }, { "literal": "Haichen Zhang" }, { "literal": "Huamin Chen" } ], "citation-key": "arxiv_2603_12823", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "274deff6cacfb5f1b7b620a86abe7ef6496b673d929de39376a9ad42dd3c4c02", "submitted_at": "2026-03-13T09:21:25Z", "updated_at": "2026-03-13T09:21:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.12823v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.12823", "catalog_id": "arxiv:2603.12823", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-13", "identifiers": { "arxiv": "2603.12823" }, "record_updated_at": "2026-03-13" }, "genre": "preprint", "id": "arxiv:2603.12823", "issued": { "date-parts": [ [ 2026, 3, 13 ] ] }, "note": "First published: 2026-03-13; Recorded arXiv version: v1; Recorded version updated: 2026-03-13T09:21:25Z.", "title": "Adaptive Vision-Language Model Routing for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.15039v1", "archive": "arXiv", "archive_location": "2603.15039", "author": [ { "literal": "Yang Li" }, { "literal": "Yuchen Liu" }, { "literal": "Haoyu Lu" }, { "literal": "Zhiqiang Xia" }, { "literal": "Hongzhen Wang" }, { "literal": "Kaiyang Han" }, { "literal": "Changpeng Yang" }, { "literal": "Jinyang Wu" }, { "literal": "Jiaming Xu" }, { "literal": "Runyu Shi" }, { "literal": "Ying Huang" } ], "citation-key": "arxiv_2603_15039", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bf6c3df80e405968d9ae2353639f3e039d3fcf8351d89db212bda038e3dfac9b", "submitted_at": "2026-03-16T09:45:33Z", "updated_at": "2026-03-16T09:45:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.15039v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.15039", "catalog_id": "arxiv:2603.15039", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-16", "identifiers": { "arxiv": "2603.15039" }, "record_updated_at": "2026-03-16" }, "genre": "preprint", "id": "arxiv:2603.15039", "issued": { "date-parts": [ [ 2026, 3, 16 ] ] }, "note": "First published: 2026-03-16; Recorded arXiv version: v1; Recorded version updated: 2026-03-16T09:45:33Z.", "title": "GUI-CEval: A Hierarchical and Comprehensive Chinese Benchmark for Mobile GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.15714v1", "archive": "arXiv", "archive_location": "2603.15714", "author": [ { "literal": "Mateusz Dziemian" }, { "literal": "Maxwell Lin" }, { "literal": "Xiaohan Fu" }, { "literal": "Micha Nowak" }, { "literal": "Nick Winter" }, { "literal": "Eliot Jones" }, { "literal": "Andy Zou" }, { "literal": "Lama Ahmad" }, { "literal": "Kamalika Chaudhuri" }, { "literal": "Sahana Chennabasappa" }, { "literal": "Xander Davies" }, { "literal": "Lauren Deason" }, { "literal": "Benjamin L. Edelman" }, { "literal": "Tanner Emek" }, { "literal": "Ivan Evtimov" }, { "literal": "Jim Gust" }, { "literal": "Maia Hamin" }, { "literal": "Kat He" }, { "literal": "Klaudia Krawiecka" }, { "literal": "Riccardo Patana" }, { "literal": "Neil Perry" }, { "literal": "Troy Peterson" }, { "literal": "Xiangyu Qi" }, { "literal": "Javier Rando" }, { "literal": "Zifan Wang" }, { "literal": "Zihan Wang" }, { "literal": "Spencer Whitman" }, { "literal": "Eric Winsor" }, { "literal": "Arman Zharmagambetov" }, { "literal": "Matt Fredrikson" }, { "literal": "Zico Kolter" } ], "citation-key": "arxiv_2603_15714", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e7fe52247e01f331021013dbc3003b3b828e134d7706ec1f587210ad286eee1d", "submitted_at": "2026-03-16T14:49:36Z", "updated_at": "2026-03-16T14:49:36Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.15714v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.15714", "catalog_id": "arxiv:2603.15714", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-16", "identifiers": { "arxiv": "2603.15714" }, "record_updated_at": "2026-03-16" }, "genre": "preprint", "id": "arxiv:2603.15714", "issued": { "date-parts": [ [ 2026, 3, 16 ] ] }, "note": "First published: 2026-03-16; Recorded arXiv version: v1; Recorded version updated: 2026-03-16T14:49:36Z.", "title": "How Vulnerable Are AI Agents to Indirect Prompt Injections? Insights from a Large-Scale Public Competition", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.17357v1", "archive": "arXiv", "archive_location": "2603.17357", "author": [ { "literal": "Nathan Zhao" } ], "citation-key": "arxiv_2603_17357", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e3b429d277ec776af18977a72aaeb95000a4fedf66aa89c0366ab02120d4cdbe", "submitted_at": "2026-03-18T04:41:16Z", "updated_at": "2026-03-18T04:41:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.17357v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.17357", "catalog_id": "arxiv:2603.17357", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-18", "identifiers": { "arxiv": "2603.17357" }, "record_updated_at": "2026-03-18" }, "genre": "preprint", "id": "arxiv:2603.17357", "issued": { "date-parts": [ [ 2026, 3, 18 ] ] }, "note": "First published: 2026-03-18; Recorded arXiv version: v1; Recorded version updated: 2026-03-18T04:41:16Z.", "title": "WebPII: Benchmarking Visual PII Detection for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.18429v1", "archive": "arXiv", "archive_location": "2603.18429", "author": [ { "literal": "Yibo Shi" }, { "literal": "Jungang Li" }, { "literal": "Linghao Zhang" }, { "literal": "Zihao Dongfang" }, { "literal": "Biao Wu" }, { "literal": "Sicheng Tao" }, { "literal": "Yibo Yan" }, { "literal": "Chenxi Qin" }, { "literal": "Weiting Liu" }, { "literal": "Zhixin Lin" }, { "literal": "Hanqian Li" }, { "literal": "Yu Huang" }, { "literal": "Song Dai" }, { "literal": "Yonghua Hei" }, { "literal": "Yue Ding" }, { "literal": "Xiang Li" }, { "literal": "Shikang Wang" }, { "literal": "Chengdong Xu" }, { "literal": "Jingqi Liu" }, { "literal": "Xueying Ma" }, { "literal": "Zhiwen Zheng" }, { "literal": "Xiaofei Zhang" }, { "literal": "Bincheng Wang" }, { "literal": "Nichen Yang" }, { "literal": "Jie Wu" }, { "literal": "Lihua Tian" }, { "literal": "Chen Li" }, { "literal": "Xuming Hu" } ], "citation-key": "arxiv_2603_18429", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0ce6da91e59ac57fa24ff991ca023b84448c54d477cacb2d6fded9bf527dd29d", "submitted_at": "2026-03-19T02:45:21Z", "updated_at": "2026-03-19T02:45:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.18429v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.18429", "catalog_id": "arxiv:2603.18429", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-19", "identifiers": { "arxiv": "2603.18429" }, "record_updated_at": "2026-03-19" }, "genre": "preprint", "id": "arxiv:2603.18429", "issued": { "date-parts": [ [ 2026, 3, 19 ] ] }, "note": "First published: 2026-03-19; Recorded arXiv version: v1; Recorded version updated: 2026-03-19T02:45:21Z.", "title": "AndroTMem: From Interaction Trajectories to Anchored Memory in Long-Horizon GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.18762v1", "archive": "arXiv", "archive_location": "2603.18762", "author": [ { "literal": "Haochen Zhao" }, { "literal": "Shaoyang Cui" } ], "citation-key": "arxiv_2603_18762", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0beb3fe342f080f0727181aba55fc1167d2512dd04542986448c5006b6b8213b", "submitted_at": "2026-03-19T11:14:45Z", "updated_at": "2026-03-19T11:14:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.18762v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.18762", "catalog_id": "arxiv:2603.18762", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-19", "identifiers": { "arxiv": "2603.18762" }, "record_updated_at": "2026-03-19" }, "genre": "preprint", "id": "arxiv:2603.18762", "issued": { "date-parts": [ [ 2026, 3, 19 ] ] }, "note": "First published: 2026-03-19; Recorded arXiv version: v1; Recorded version updated: 2026-03-19T11:14:45Z.", "title": "ClawTrap: A MITM-Based Red-Teaming Framework for Real-World OpenClaw Security Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.19191v1", "archive": "arXiv", "archive_location": "2603.19191", "author": [ { "literal": "Zehao Li" }, { "literal": "Zhenyu Wu" }, { "literal": "Yibo Zhao" }, { "literal": "Bowen Yang" }, { "literal": "Jingjing Xie" }, { "literal": "Zhaoyang Liu" }, { "literal": "Zhoumianze Liu" }, { "literal": "Kaiming Jin" }, { "literal": "Jianze Liang" }, { "literal": "Zonglin Li" }, { "literal": "Feng Wu" }, { "literal": "Bowen Zhou" }, { "literal": "Zun Wang" }, { "literal": "Zichen Ding" } ], "citation-key": "arxiv_2603_19191", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ff31371580af6a0c43a6c26f076d5060ee06f974d3a6c1ade67348dc0af4859d", "submitted_at": "2026-03-19T17:47:47Z", "updated_at": "2026-03-19T17:47:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.19191v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.19191", "catalog_id": "arxiv:2603.19191", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-19", "identifiers": { "arxiv": "2603.19191" }, "record_updated_at": "2026-03-19" }, "genre": "preprint", "id": "arxiv:2603.19191", "issued": { "date-parts": [ [ 2026, 3, 19 ] ] }, "note": "First published: 2026-03-19; Recorded arXiv version: v1; Recorded version updated: 2026-03-19T17:47:47Z.", "title": "OS-Themis: A Scalable Critic Framework for Generalist GUI Rewards", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.19685v1", "archive": "arXiv", "archive_location": "2603.19685", "author": [ { "literal": "Taiyi Wang" }, { "literal": "Sian Gooding" }, { "literal": "Florian Hartmann" }, { "literal": "Oriana Riva" }, { "literal": "Edward Grefenstette" } ], "citation-key": "arxiv_2603_19685", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5538933a0e3d11df6ff4d767849979703d83412599fbbb0f8b432c4d9850633b", "submitted_at": "2026-03-20T06:32:16Z", "updated_at": "2026-03-20T06:32:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.19685v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.19685", "catalog_id": "arxiv:2603.19685", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-20", "identifiers": { "arxiv": "2603.19685" }, "record_updated_at": "2026-03-20" }, "genre": "preprint", "id": "arxiv:2603.19685", "issued": { "date-parts": [ [ 2026, 3, 20 ] ] }, "note": "First published: 2026-03-20; Recorded arXiv version: v1; Recorded version updated: 2026-03-20T06:32:16Z.", "title": "A Subgoal-driven Framework for Improving Long-Horizon LLM Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.19993v1", "archive": "arXiv", "archive_location": "2603.19993", "author": [ { "literal": "Rozain Shakeel" }, { "literal": "Abdul Rahman Mohammad Ali" }, { "literal": "Muneeb Mushtaq" }, { "literal": "Tausifa Jan Saleem" }, { "literal": "Tajamul Ashraf" } ], "citation-key": "arxiv_2603_19993", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cfa6cd9aead44150fdc997654ded7253b2890e936b94e44be2fb33642b24d302", "submitted_at": "2026-03-20T14:43:53Z", "updated_at": "2026-03-20T14:43:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.19993v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.19993", "catalog_id": "arxiv:2603.19993", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-20", "identifiers": { "arxiv": "2603.19993" }, "record_updated_at": "2026-03-20" }, "genre": "preprint", "id": "arxiv:2603.19993", "issued": { "date-parts": [ [ 2026, 3, 20 ] ] }, "note": "First published: 2026-03-20; Recorded arXiv version: v1; Recorded version updated: 2026-03-20T14:43:53Z.", "title": "MedSPOT: A Workflow-Aware Sequential Grounding Benchmark for Clinical GUI", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.22529v1", "archive": "arXiv", "archive_location": "2603.22529", "author": [ { "literal": "Shoubin Yu" }, { "literal": "Lei Shu" }, { "literal": "Antoine Yang" }, { "literal": "Yao Fu" }, { "literal": "Srinivas Sunkara" }, { "literal": "Maria Wang" }, { "literal": "Jindong Chen" }, { "literal": "Mohit Bansal" }, { "literal": "Boqing Gong" } ], "citation-key": "arxiv_2603_22529", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3e4ed7391969359c2594e8b96b8a3a559ef98d2c84a10c1a2c39b3bd962b58c5", "submitted_at": "2026-03-23T19:47:58Z", "updated_at": "2026-03-23T19:47:58Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.22529v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.22529", "catalog_id": "arxiv:2603.22529", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-23", "identifiers": { "arxiv": "2603.22529" }, "record_updated_at": "2026-03-23" }, "genre": "preprint", "id": "arxiv:2603.22529", "issued": { "date-parts": [ [ 2026, 3, 23 ] ] }, "note": "First published: 2026-03-23; Recorded arXiv version: v1; Recorded version updated: 2026-03-23T19:47:58Z.", "title": "Ego2Web: A Web Agent Benchmark Grounded in Egocentric Videos", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.23007v1", "archive": "arXiv", "archive_location": "2603.23007", "author": [ { "literal": "Yutao Luo" }, { "literal": "Haotian Zhu" }, { "literal": "Shuchao Pang" }, { "literal": "Zhigang Lu" }, { "literal": "Tian Dong" }, { "literal": "Yongbin Zhou" }, { "literal": "Minhui Xue" } ], "citation-key": "arxiv_2603_23007", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "37960f42162d2bff2fca5e71e350f1ef7c4d86e3f438fd777a88dc2280d3aabe", "submitted_at": "2026-03-24T09:51:43Z", "updated_at": "2026-03-24T09:51:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.23007v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.23007", "catalog_id": "arxiv:2603.23007", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-24", "identifiers": { "arxiv": "2603.23007" }, "record_updated_at": "2026-03-24" }, "genre": "preprint", "id": "arxiv:2603.23007", "issued": { "date-parts": [ [ 2026, 3, 24 ] ] }, "note": "First published: 2026-03-24; Recorded arXiv version: v1; Recorded version updated: 2026-03-24T09:51:43Z.", "title": "AgentRAE: Remote Action Execution through Notification-based Visual Backdoors against Screenshots-based Mobile GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.24157v1", "archive": "arXiv", "archive_location": "2603.24157", "author": [ { "literal": "Akash Ghosh" }, { "literal": "Tajamul Ashraf" }, { "literal": "Rishu Kumar Singh" }, { "literal": "Numan Saeed" }, { "literal": "Sriparna Saha" }, { "literal": "Xiuying Chen" }, { "literal": "Salman Khan" } ], "citation-key": "arxiv_2603_24157", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8fb6a294616e1be46c5da05f60cda7d43c05d8a1ce71116c5e30cb94507cd56f", "submitted_at": "2026-03-25T10:25:48Z", "updated_at": "2026-03-25T10:25:48Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.24157v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.24157", "catalog_id": "arxiv:2603.24157", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-25", "identifiers": { "arxiv": "2603.24157" }, "record_updated_at": "2026-03-25" }, "genre": "preprint", "id": "arxiv:2603.24157", "issued": { "date-parts": [ [ 2026, 3, 25 ] ] }, "note": "First published: 2026-03-25; Recorded arXiv version: v1; Recorded version updated: 2026-03-25T10:25:48Z.", "title": "CarePilot: A Multi-Agent Framework for Long-Horizon Computer Task Automation in Healthcare", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.24440v1", "archive": "arXiv", "archive_location": "2603.24440", "author": [ { "literal": "Xiangru Jian" }, { "literal": "Shravan Nayak" }, { "literal": "Kevin Qinghong Lin" }, { "literal": "Aarash Feizi" }, { "literal": "Kaixin Li" }, { "literal": "Patrice Bechard" }, { "literal": "Spandana Gella" }, { "literal": "Sai Rajeswar" } ], "citation-key": "arxiv_2603_24440", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "15237fda2f3799314009635017487749dbedc9b1a0ffdd6b56a8d28eaf01481d", "submitted_at": "2026-03-25T15:52:56Z", "updated_at": "2026-03-25T15:52:56Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.24440v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.24440", "catalog_id": "arxiv:2603.24440", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-25", "identifiers": { "arxiv": "2603.24440" }, "record_updated_at": "2026-03-25" }, "genre": "preprint", "id": "arxiv:2603.24440", "issued": { "date-parts": [ [ 2026, 3, 25 ] ] }, "note": "First published: 2026-03-25; Recorded arXiv version: v1; Recorded version updated: 2026-03-25T15:52:56Z.", "title": "CUA-Suite: Massive Human-annotated Video Demonstrations for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.25226v1", "archive": "arXiv", "archive_location": "2603.25226", "author": [ { "literal": "Fanheng Kong" }, { "literal": "Jingyuan Zhang" }, { "literal": "Yang Yue" }, { "literal": "Chenxi Sun" }, { "literal": "Yang Tian" }, { "literal": "Shi Feng" }, { "literal": "Xiaocui Yang" }, { "literal": "Daling Wang" }, { "literal": "Yu Tian" }, { "literal": "Jun Du" }, { "literal": "Wenchong Zeng" }, { "literal": "Han Li" }, { "literal": "Kun Gai" } ], "citation-key": "arxiv_2603_25226", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8414eb53c66f745fb074d9f2699f76ffea81cb55f98aacf9ce896401937f80c5", "submitted_at": "2026-03-26T09:27:29Z", "updated_at": "2026-03-26T09:27:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.25226v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.25226", "catalog_id": "arxiv:2603.25226", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-26", "identifiers": { "arxiv": "2603.25226" }, "record_updated_at": "2026-03-26" }, "genre": "preprint", "id": "arxiv:2603.25226", "issued": { "date-parts": [ [ 2026, 3, 26 ] ] }, "note": "First published: 2026-03-26; Recorded arXiv version: v1; Recorded version updated: 2026-03-26T09:27:29Z.", "title": "WebTestBench: Evaluating Computer-Use Agents towards End-to-End Automated Web Testing", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.25864v1", "archive": "arXiv", "archive_location": "2603.25864", "author": [ { "literal": "Saelyne Yang" }, { "literal": "Jaesang Yu" }, { "literal": "Yi-Hao Peng" }, { "literal": "Kevin Qinghong Lin" }, { "literal": "Jae Won Cho" }, { "literal": "Yale Song" }, { "literal": "Juho Kim" } ], "citation-key": "arxiv_2603_25864", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9b3dd3efca196fa3dd168431102dbb5f0f763e7d9e54b3d709ad2362a9c6f154", "submitted_at": "2026-03-26T19:37:53Z", "updated_at": "2026-03-26T19:37:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.25864v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.25864", "catalog_id": "arxiv:2603.25864", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-26", "identifiers": { "arxiv": "2603.25864" }, "record_updated_at": "2026-03-26" }, "genre": "preprint", "id": "arxiv:2603.25864", "issued": { "date-parts": [ [ 2026, 3, 26 ] ] }, "note": "First published: 2026-03-26; Recorded arXiv version: v1; Recorded version updated: 2026-03-26T19:37:53Z.", "title": "GUIDE: A Benchmark for Understanding and Assisting Users in Open-Ended GUI Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.29020v1", "archive": "arXiv", "archive_location": "2603.29020", "author": [ { "literal": "Deepak Akkil" }, { "literal": "Mowafak Allaham" }, { "literal": "Amal Raj" }, { "literal": "Tamer Abuelsaad" }, { "literal": "Ravi Kokku" } ], "citation-key": "arxiv_2603_29020", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b4f03e65387e3aad99b8c4cef166a7a8b4a9e2954ad932b8daf566e30f156595", "submitted_at": "2026-03-30T21:27:28Z", "updated_at": "2026-03-30T21:27:28Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.29020v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.29020", "catalog_id": "arxiv:2603.29020", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-30", "identifiers": { "arxiv": "2603.29020" }, "record_updated_at": "2026-03-30" }, "genre": "preprint", "id": "arxiv:2603.29020", "issued": { "date-parts": [ [ 2026, 3, 30 ] ] }, "note": "First published: 2026-03-30; Recorded arXiv version: v1; Recorded version updated: 2026-03-30T21:27:28Z.", "title": "Emergence WebVoyager: Toward Consistent and Transparent Evaluation of (Web) Agents in The Wild", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2603.29318v1", "archive": "arXiv", "archive_location": "2603.29318", "author": [ { "literal": "Hongyi Nie" }, { "literal": "Xunyuan Liu" }, { "literal": "Yudong Bai" }, { "literal": "Yaqing Wang" }, { "literal": "Yang Liu" }, { "literal": "Quanming Yao" }, { "literal": "Zhen Wang" } ], "citation-key": "arxiv_2603_29318", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1c71b713efd6ea47e7c78f5dd51d06c5cfdba750529bfbcdfc8e1c69897cbcc9", "submitted_at": "2026-03-31T06:37:12Z", "updated_at": "2026-03-31T06:37:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2603.29318v1" }, "canonical_source_url": "https://arxiv.org/abs/2603.29318", "catalog_id": "arxiv:2603.29318", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-03-31", "identifiers": { "arxiv": "2603.29318" }, "record_updated_at": "2026-03-31" }, "genre": "preprint", "id": "arxiv:2603.29318", "issued": { "date-parts": [ [ 2026, 3, 31 ] ] }, "note": "First published: 2026-03-31; Recorded arXiv version: v1; Recorded version updated: 2026-03-31T06:37:12Z.", "title": "PSPA-Bench: A Personalized Benchmark for Smartphone GUI Agent", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.00886v1", "archive": "arXiv", "archive_location": "2604.00886", "author": [ { "literal": "Nan Wang" }, { "literal": "Zhiwei Jin" }, { "literal": "Chen Chen" }, { "literal": "Haonan Lu" } ], "citation-key": "arxiv_2604_00886", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ce7f70ff1f8c7d4058d8fdf00cb6cf1d61c0367a50ba76fc96d356bb07452c60", "submitted_at": "2026-04-01T13:33:27Z", "updated_at": "2026-04-01T13:33:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.00886v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.00886", "catalog_id": "arxiv:2604.00886", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-04-01", "identifiers": { "arxiv": "2604.00886" }, "record_updated_at": "2026-04-01" }, "genre": "preprint", "id": "arxiv:2604.00886", "issued": { "date-parts": [ [ 2026, 4, 1 ] ] }, "note": "First published: 2026-04-01; Recorded arXiv version: v1; Recorded version updated: 2026-04-01T13:33:27Z.", "title": "PixelPrune: Pixel-Level Adaptive Visual Token Reduction via Predictive Coding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.00986v2", "archive": "arXiv", "archive_location": "2604.00986", "author": [ { "literal": "Zhengyang Tang" }, { "literal": "Ke Ji" }, { "literal": "Xidong Wang" }, { "literal": "Zihan Ye" }, { "literal": "Xinyuan Wang" }, { "literal": "Yiduo Guo" }, { "literal": "Ziniu Li" }, { "literal": "Chenxin Li" }, { "literal": "Jingyuan Hu" }, { "literal": "Shunian Chen" }, { "literal": "Tongxu Luo" }, { "literal": "Jiaxi Bi" }, { "literal": "Zeyu Qin" }, { "literal": "Shaobo Wang" }, { "literal": "Xin Lai" }, { "literal": "Pengyuan Lyu" }, { "literal": "Junyi Li" }, { "literal": "Can Xu" }, { "literal": "Chengquan Zhang" }, { "literal": "Han Hu" }, { "literal": "Ming Yan" }, { "literal": "Benyou Wang" } ], "citation-key": "arxiv_2604_00986", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3f137028b7608598fb1885a0bf1330e00c9404851571f856665aae1af1d9d63c", "submitted_at": "2026-04-01T14:50:50Z", "updated_at": "2026-04-02T13:51:55Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.00986v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.00986", "catalog_id": "arxiv:2604.00986", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-01", "identifiers": { "arxiv": "2604.00986" }, "record_updated_at": "2026-04-02" }, "genre": "preprint", "id": "arxiv:2604.00986", "issued": { "date-parts": [ [ 2026, 4, 1 ] ] }, "note": "First published: 2026-04-01; Recorded arXiv version: v2; Recorded version updated: 2026-04-02T13:51:55Z.", "title": "Do Phone-Use Agents Respect Your Privacy?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.02345v1", "archive": "arXiv", "archive_location": "2604.02345", "author": [ { "literal": "Mengzhou Wu" }, { "literal": "Yuzhe Guo" }, { "literal": "Yuan Cao" }, { "literal": "Haochuan Lu" }, { "literal": "Songhe Zhu" }, { "literal": "Pingzhe Qu" }, { "literal": "Xin Chen" }, { "literal": "Kang Qin" }, { "literal": "Zhongpu Wang" }, { "literal": "Xiaode Zhang" }, { "literal": "Xinyi Wang" }, { "literal": "Wei Dai" }, { "literal": "Gang Cao" }, { "literal": "Yuetang Deng" }, { "literal": "Zhi Gong" }, { "literal": "Dezhi Ran" }, { "literal": "Linyi Li" }, { "literal": "Wei Yang" }, { "literal": "Tao Xie" } ], "citation-key": "arxiv_2604_02345", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bac86c2ff2bf18feb3edfbe76285d33718d0c6d9564aff5a9d62f8778cdc5147", "submitted_at": "2026-02-11T17:02:38Z", "updated_at": "2026-02-11T17:02:38Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.02345v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.02345", "catalog_id": "arxiv:2604.02345", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-11", "identifiers": { "arxiv": "2604.02345" }, "record_updated_at": "2026-02-11" }, "genre": "preprint", "id": "arxiv:2604.02345", "issued": { "date-parts": [ [ 2026, 2, 11 ] ] }, "note": "First published: 2026-02-11; Recorded arXiv version: v1; Recorded version updated: 2026-02-11T17:02:38Z.", "title": "UI-Oceanus: Scaling GUI Agents with Synthetic Environmental Dynamics", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.04399v1", "archive": "arXiv", "archive_location": "2604.04399", "author": [ { "literal": "Yuwen Zhai" }, { "literal": "Runze Li" }, { "literal": "Liang Wang" }, { "literal": "Nian Shi" }, { "literal": "Liwu Xu" }, { "literal": "Wei Zhang" }, { "literal": "Ran Lin" }, { "literal": "Bo Xu" }, { "literal": "Benlei Cui" } ], "citation-key": "arxiv_2604_04399", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "856e9920aea7d821061b3258b31aa32fd37b5579d628f5507f45ba83f237b77b", "submitted_at": "2026-04-06T03:58:43Z", "updated_at": "2026-04-06T03:58:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.04399v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.04399", "catalog_id": "arxiv:2604.04399", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-06", "identifiers": { "arxiv": "2604.04399" }, "record_updated_at": "2026-04-06" }, "genre": "preprint", "id": "arxiv:2604.04399", "issued": { "date-parts": [ [ 2026, 4, 6 ] ] }, "note": "First published: 2026-04-06; Recorded arXiv version: v1; Recorded version updated: 2026-04-06T03:58:43Z.", "title": "GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.04918v2", "archive": "arXiv", "archive_location": "2604.04918", "author": [ { "literal": "Chaoran Chen" }, { "literal": "Zhiping Zhang" }, { "literal": "Zeya Chen" }, { "literal": "Eryue Xu" }, { "literal": "Yinuo Yang" }, { "literal": "Ibrahim Khalilov" }, { "literal": "Simret A Gebreegziabher" }, { "literal": "Yanfang Ye" }, { "literal": "Ziang Xiao" }, { "literal": "Yaxing Yao" }, { "literal": "Tianshi Li" }, { "literal": "Toby Jia-Jun Li" } ], "citation-key": "arxiv_2604_04918", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1411ee71ad8f65e4941af4ef538f52da6e9bf9c57fb537b6ecf37c8478609693", "submitted_at": "2026-04-06T17:57:16Z", "updated_at": "2026-09-12T06:05:52Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.04918v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.04918", "catalog_id": "arxiv:2604.04918", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-06", "identifiers": { "arxiv": "2604.04918" }, "record_updated_at": "2026-09-12" }, "genre": "preprint", "id": "arxiv:2604.04918", "issued": { "date-parts": [ [ 2026, 4, 6 ] ] }, "note": "First published: 2026-04-06; Recorded arXiv version: v2; Recorded version updated: 2026-09-12T06:05:52Z.", "title": "Comparing Human Oversight Strategies for Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.05157v3", "archive": "arXiv", "archive_location": "2604.05157", "author": [ { "literal": "Rongqian Chen" }, { "literal": "Yu Li" }, { "literal": "Zeyu Fang" }, { "literal": "Sizhe Tang" }, { "literal": "Weidong Cao" }, { "literal": "Tian Lan" } ], "citation-key": "arxiv_2604_05157", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "238aac264fb21022f7c546ae3aa3f4ea104c381f5d318f6d962f22fb00095216", "submitted_at": "2026-04-06T20:39:30Z", "updated_at": "2026-05-28T08:13:55Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2604.05157v3" }, "canonical_source_url": "https://arxiv.org/abs/2604.05157", "catalog_id": "arxiv:2604.05157", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-06", "identifiers": { "arxiv": "2604.05157" }, "record_updated_at": "2026-05-28" }, "genre": "preprint", "id": "arxiv:2604.05157", "issued": { "date-parts": [ [ 2026, 4, 6 ] ] }, "note": "First published: 2026-04-06; Recorded arXiv version: v3; Recorded version updated: 2026-05-28T08:13:55Z.", "title": "IntentScore: Intent-Conditioned Action Evaluation for Computer-Use Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2604.06126v1", "archive": "arXiv", "archive_location": "2604.06126", "author": [ { "literal": "Pranjal Aggarwal" }, { "literal": "Graham Neubig" }, { "literal": "Sean Welleck" } ], "citation-key": "arxiv_2604_06126", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fdd9cb6daad8832261ebe861fc1cb74429b819305e14ecfeb9698616013e6ac8", "submitted_at": "2026-04-07T17:38:15Z", "updated_at": "2026-04-07T17:38:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.06126v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.06126", "catalog_id": "arxiv:2604.06126", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-07", "identifiers": { "arxiv": "2604.06126" }, "record_updated_at": "2026-04-07" }, "genre": "preprint", "id": "arxiv:2604.06126", "issued": { "date-parts": [ [ 2026, 4, 7 ] ] }, "note": "First published: 2026-04-07; Recorded arXiv version: v1; Recorded version updated: 2026-04-07T17:38:15Z.", "title": "Gym-Anything: Turn any Software into an Agent Environment", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.06240v1", "archive": "arXiv", "archive_location": "2604.06240", "author": [ { "literal": "Corby Rosset" }, { "literal": "Pratyusha Sharma" }, { "literal": "Andrew Zhao" }, { "literal": "Miguel Gonzalez-Fernandez" }, { "literal": "Ahmed Awadallah" } ], "citation-key": "arxiv_2604_06240", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b452684b305c4f20ded686289b448fbe4f08bb3b34f515a43e2ce5d6dc4ef506", "submitted_at": "2026-04-05T06:26:05Z", "updated_at": "2026-04-05T06:26:05Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.06240v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.06240", "catalog_id": "arxiv:2604.06240", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-05", "identifiers": { "arxiv": "2604.06240" }, "record_updated_at": "2026-04-05" }, "genre": "preprint", "id": "arxiv:2604.06240", "issued": { "date-parts": [ [ 2026, 4, 5 ] ] }, "note": "First published: 2026-04-05; Recorded arXiv version: v1; Recorded version updated: 2026-04-05T06:26:05Z.", "title": "The Art of Building Verifiers for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.06367v2", "archive": "arXiv", "archive_location": "2604.06367", "author": [ { "literal": "Guruprasad Viswanathan Ramesh" }, { "literal": "Asmit Nayak" }, { "literal": "Basieem Siddique" }, { "literal": "Kassem Fawaz" } ], "citation-key": "arxiv_2604_06367", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "400d8a9f7aed4f150dd5198d9a1561d2df0454069495594ca165ad4254a1be59", "submitted_at": "2026-04-07T18:43:21Z", "updated_at": "2026-06-16T22:37:53Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.06367v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.06367", "catalog_id": "arxiv:2604.06367", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-07", "identifiers": { "arxiv": "2604.06367" }, "record_updated_at": "2026-06-16" }, "genre": "preprint", "id": "arxiv:2604.06367", "issued": { "date-parts": [ [ 2026, 4, 7 ] ] }, "note": "First published: 2026-04-07; Recorded arXiv version: v2; Recorded version updated: 2026-06-16T22:37:53Z.", "title": "WebSP-Eval: Evaluating Web Agents on Website Security and Privacy Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.06995v3", "archive": "arXiv", "archive_location": "2604.06995", "author": [ { "literal": "Songze Li" }, { "literal": "Xiaoke Guo" }, { "literal": "Tianqi Liu" }, { "literal": "Biao Yi" }, { "literal": "Zhaoyan Gong" }, { "literal": "Zhiqiang Liu" }, { "literal": "Huajun Chen" }, { "literal": "Wen Zhang" } ], "citation-key": "arxiv_2604_06995", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b99643e43c5e20af088d6fa62152e03256c947af76a6b3c0abfc6f74a0f7d648", "submitted_at": "2026-04-08T12:12:09Z", "updated_at": "2026-10-02T06:30:10Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2604.06995v3" }, "canonical_source_url": "https://arxiv.org/abs/2604.06995", "catalog_id": "arxiv:2604.06995", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-08", "identifiers": { "arxiv": "2604.06995" }, "record_updated_at": "2026-10-02" }, "genre": "preprint", "id": "arxiv:2604.06995", "issued": { "date-parts": [ [ 2026, 4, 8 ] ] }, "note": "First published: 2026-04-08; Recorded arXiv version: v3; Recorded version updated: 2026-10-02T06:30:10Z.", "title": "What's Missing in Screen-to-Action? Towards a UI-in-the-Loop Paradigm for Multimodal GUI Reasoning", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2604.07831v2", "archive": "arXiv", "archive_location": "2604.07831", "author": [ { "literal": "Wenkui Yang" }, { "literal": "Chao Jin" }, { "literal": "Haisu Zhu" }, { "literal": "Weilin Luo" }, { "literal": "Derek Yuen" }, { "literal": "Kun Shao" }, { "literal": "Junxian Duan" }, { "literal": "Huaibo Huang" }, { "literal": "Jie Cao" }, { "literal": "Ran He" } ], "citation-key": "arxiv_2604_07831", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a5427eedff0e97f9c73101403754c38e8928ad71abab7b7eff5f136a30ee2205", "submitted_at": "2026-04-09T05:32:34Z", "updated_at": "2026-07-08T02:39:35Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.07831v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.07831", "catalog_id": "arxiv:2604.07831", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-09", "identifiers": { "arxiv": "2604.07831" }, "record_updated_at": "2026-07-08" }, "genre": "preprint", "id": "arxiv:2604.07831", "issued": { "date-parts": [ [ 2026, 4, 9 ] ] }, "note": "First published: 2026-04-09; Recorded arXiv version: v2; Recorded version updated: 2026-07-08T02:39:35Z.", "title": "Are GUI Agents Focused Enough? Automated Distraction via Semantic-level UI Element Injection", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.09155v1", "archive": "arXiv", "archive_location": "2604.09155", "author": [ { "literal": "Yushi Feng" }, { "literal": "Junye Du" }, { "literal": "Qifan Wang" }, { "literal": "Zizhan Ma" }, { "literal": "Qian Niu" }, { "literal": "Yutaka Matsuo" }, { "literal": "Long Feng" }, { "literal": "Lequan Yu" } ], "citation-key": "arxiv_2604_09155", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e53647528b9ee9d98f93ee665d0d886f9b400a129873e240704041df291f5749", "submitted_at": "2026-04-10T09:41:21Z", "updated_at": "2026-04-10T09:41:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.09155v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.09155", "catalog_id": "arxiv:2604.09155", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-10", "identifiers": { "arxiv": "2604.09155" }, "record_updated_at": "2026-04-10" }, "genre": "preprint", "id": "arxiv:2604.09155", "issued": { "date-parts": [ [ 2026, 4, 10 ] ] }, "note": "First published: 2026-04-10; Recorded arXiv version: v1; Recorded version updated: 2026-04-10T09:41:21Z.", "title": "CORA: Conformal Risk-Controlled Agents for Safeguarded Mobile GUI Automation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.09581v2", "archive": "arXiv", "archive_location": "2604.09581", "author": [ { "literal": "Wee Joe Tan" }, { "literal": "Zi Rui Lucas Lim" }, { "literal": "Shashank Durgad" }, { "literal": "Karim Obegi" }, { "literal": "Aiden Yiliu Li" } ], "citation-key": "arxiv_2604_09581", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "445e520a60e0e3cc0c3c519e21bfa8eebfa33d4fe4320c437e67b94a02d84504", "submitted_at": "2026-02-25T18:59:42Z", "updated_at": "2026-04-14T20:54:46Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.09581v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.09581", "catalog_id": "arxiv:2604.09581", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-02-25", "identifiers": { "arxiv": "2604.09581" }, "record_updated_at": "2026-04-14" }, "genre": "preprint", "id": "arxiv:2604.09581", "issued": { "date-parts": [ [ 2026, 2, 25 ] ] }, "note": "First published: 2026-02-25; Recorded arXiv version: v2; Recorded version updated: 2026-04-14T20:54:46Z.", "title": "Avenir-UX: Automated UX Evaluation via Simulated Human Web Interaction with GUI Grounding", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.09937v1", "archive": "arXiv", "archive_location": "2604.09937", "author": [ { "literal": "Suhana Bedi" }, { "literal": "Ryan Welch" }, { "literal": "Ethan Steinberg" }, { "literal": "Michael Wornow" }, { "literal": "Taeil Matthew Kim" }, { "literal": "Haroun Ahmed" }, { "literal": "Peter Sterling" }, { "literal": "Bravim Purohit" }, { "literal": "Qurat Akram" }, { "literal": "Angelic Acosta" }, { "literal": "Esther Nubla" }, { "literal": "Pritika Sharma" }, { "literal": "Michael A. Pfeffer" }, { "literal": "Sanmi Koyejo" }, { "literal": "Nigam H. Shah" } ], "citation-key": "arxiv_2604_09937", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "55cedd4ad9760618fe9b3354e384f69b69b23b65f75837a3898484c9c6e55586", "submitted_at": "2026-04-10T22:33:39Z", "updated_at": "2026-04-10T22:33:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.09937v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.09937", "catalog_id": "arxiv:2604.09937", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-10", "identifiers": { "arxiv": "2604.09937" }, "record_updated_at": "2026-04-10" }, "genre": "preprint", "id": "arxiv:2604.09937", "issued": { "date-parts": [ [ 2026, 4, 10 ] ] }, "note": "First published: 2026-04-10; Recorded arXiv version: v1; Recorded version updated: 2026-04-10T22:33:39Z.", "title": "HealthAdminBench: Evaluating Computer-Use Agents on Healthcare Administration Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.10988v1", "archive": "arXiv", "archive_location": "2604.10988", "author": [ { "literal": "Peng Yuan" }, { "literal": "Yuyang Yin" }, { "literal": "Yuxuan Cai" }, { "literal": "Zheng Wei" } ], "citation-key": "arxiv_2604_10988", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1db8d789fb8ba2d5e61ca24536670fd83491c305f532adfecf7193821f7eea52", "submitted_at": "2026-04-13T04:45:27Z", "updated_at": "2026-04-13T04:45:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.10988v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.10988", "catalog_id": "arxiv:2604.10988", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-13", "identifiers": { "arxiv": "2604.10988" }, "record_updated_at": "2026-04-13" }, "genre": "preprint", "id": "arxiv:2604.10988", "issued": { "date-parts": [ [ 2026, 4, 13 ] ] }, "note": "First published: 2026-04-13; Recorded arXiv version: v1; Recorded version updated: 2026-04-13T04:45:27Z.", "title": "WebForge: Breaking the Realism-Reproducibility-Scalability Trilemma in Browser Agent Benchmark", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.11784v1", "archive": "arXiv", "archive_location": "2604.11784", "author": [ { "literal": "Fei Tang" }, { "literal": "Zhiqiong Lu" }, { "literal": "Boxuan Zhang" }, { "literal": "Weiming Lu" }, { "literal": "Jun Xiao" }, { "literal": "Yueting Zhuang" }, { "literal": "Yongliang Shen" } ], "citation-key": "arxiv_2604_11784", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "df56b38203bc5ec8d7e871eb2ad9fca9891a049df62b7d72796fe400eb8bce5e", "submitted_at": "2026-04-13T17:52:04Z", "updated_at": "2026-04-13T17:52:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.11784v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.11784", "catalog_id": "arxiv:2604.11784", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-13", "identifiers": { "arxiv": "2604.11784" }, "record_updated_at": "2026-04-13" }, "genre": "preprint", "id": "arxiv:2604.11784", "issued": { "date-parts": [ [ 2026, 4, 13 ] ] }, "note": "First published: 2026-04-13; Recorded arXiv version: v1; Recorded version updated: 2026-04-13T17:52:04Z.", "title": "ClawGUI: A Unified Framework for Training, Evaluating, and Deploying GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.12284v1", "archive": "arXiv", "archive_location": "2604.12284", "author": [ { "literal": "Yulin Chen" }, { "literal": "Tri Cao" }, { "literal": "Haoran Li" }, { "literal": "Yue Liu" }, { "literal": "Yibo Li" }, { "literal": "Yufei He" }, { "literal": "Le Minh Khoi" }, { "literal": "Yangqiu Song" }, { "literal": "Shuicheng Yan" }, { "literal": "Bryan Hooi" } ], "citation-key": "arxiv_2604_12284", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "64d4323050a5ad0f7265231337eb22a4c0823cee41587099c8bdc5db6b320dea", "submitted_at": "2026-04-14T04:50:35Z", "updated_at": "2026-04-14T04:50:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.12284v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.12284", "catalog_id": "arxiv:2604.12284", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-14", "identifiers": { "arxiv": "2604.12284" }, "record_updated_at": "2026-04-14" }, "genre": "preprint", "id": "arxiv:2604.12284", "issued": { "date-parts": [ [ 2026, 4, 14 ] ] }, "note": "First published: 2026-04-14; Recorded arXiv version: v1; Recorded version updated: 2026-04-14T04:50:35Z.", "title": "WebAgentGuard: A Reasoning-Driven Guard Model for Detecting Prompt Injection Attacks in Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.14262v2", "archive": "arXiv", "archive_location": "2604.14262", "author": [ { "literal": "Yangyue Wang" }, { "literal": "Harshvardhan Sikka" }, { "literal": "Yash Mathur" }, { "literal": "Tony Zhou" }, { "literal": "Jinu Nyachhyon" }, { "literal": "Pranav Guruprasad" } ], "citation-key": "arxiv_2604_14262", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5dc36ff0c9c4bd39de360824967e53c18fe57562cb8da1462d3e63c0dd348e71", "submitted_at": "2026-04-15T16:39:22Z", "updated_at": "2026-07-01T10:12:28Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.14262v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.14262", "catalog_id": "arxiv:2604.14262", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-15", "identifiers": { "arxiv": "2604.14262" }, "record_updated_at": "2026-07-01" }, "genre": "preprint", "id": "arxiv:2604.14262", "issued": { "date-parts": [ [ 2026, 4, 15 ] ] }, "note": "First published: 2026-04-15; Recorded arXiv version: v2; Recorded version updated: 2026-07-01T10:12:28Z.", "title": "GUI-Perturbed: Domain Randomization Reveals Systematic Brittleness in GUI Grounding Models", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.14668v1", "archive": "arXiv", "archive_location": "2604.14668", "author": [ { "literal": "Pan Hao" }, { "literal": "Rishi Selvakumaran" }, { "literal": "Jacob Sun" }, { "literal": "Qianwen Wang" } ], "citation-key": "arxiv_2604_14668", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bd3b5e5b9a607023c82150673255eaf2f90c22db86d3c049b2a2fd536aaef17f", "submitted_at": "2026-04-16T06:22:48Z", "updated_at": "2026-04-16T06:22:48Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.14668v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.14668", "catalog_id": "arxiv:2604.14668", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-16", "identifiers": { "arxiv": "2604.14668" }, "record_updated_at": "2026-04-16" }, "genre": "preprint", "id": "arxiv:2604.14668", "issued": { "date-parts": [ [ 2026, 4, 16 ] ] }, "note": "First published: 2026-04-16; Recorded arXiv version: v1; Recorded version updated: 2026-04-16T06:22:48Z.", "title": "Beyond Chat and Clicks: GUI Agents for In-Situ Assistance via Live Interface Transformation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.14956v1", "archive": "arXiv", "archive_location": "2604.14956", "author": [ { "literal": "Wenhao Wang" }, { "literal": "Haoting Shi" }, { "literal": "Mengying Yuan" }, { "literal": "Yiquan Lin" }, { "literal": "Panrong Tong" }, { "literal": "Hanzhang Zhou" }, { "literal": "Guangyi Liu" }, { "literal": "Pengxiang Zhao" }, { "literal": "Yue Wang" }, { "literal": "Siheng Chen" } ], "citation-key": "arxiv_2604_14956", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1e2b5d7149adcbe634df0bfdc9c34ba436e51e6891df5c4836daea9795061164", "submitted_at": "2026-04-16T12:53:03Z", "updated_at": "2026-04-16T12:53:03Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.14956v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.14956", "catalog_id": "arxiv:2604.14956", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-16", "identifiers": { "arxiv": "2604.14956" }, "record_updated_at": "2026-04-16" }, "genre": "preprint", "id": "arxiv:2604.14956", "issued": { "date-parts": [ [ 2026, 4, 16 ] ] }, "note": "First published: 2026-04-16; Recorded arXiv version: v1; Recorded version updated: 2026-04-16T12:53:03Z.", "title": "FedGUI: Benchmarking Federated GUI Agents across Heterogeneous Platforms, Devices, and Operating Systems", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.17284v1", "archive": "arXiv", "archive_location": "2604.17284", "author": [ { "literal": "Chao Jin" }, { "literal": "Wenkui Yang" }, { "literal": "Hao Sun" }, { "literal": "Yuqi Liao" }, { "literal": "Qianyi Jiang" }, { "literal": "Kai Zhou" }, { "literal": "Jie Cao" }, { "literal": "Ran He" }, { "literal": "Huaibo Huang" } ], "citation-key": "arxiv_2604_17284", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2b102c9363b1be77273c7fab6f7f5cde74157c646671c93b9503770fe60dd70c", "submitted_at": "2026-04-19T06:55:59Z", "updated_at": "2026-04-19T06:55:59Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.17284v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.17284", "catalog_id": "arxiv:2604.17284", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-19", "identifiers": { "arxiv": "2604.17284" }, "record_updated_at": "2026-04-19" }, "genre": "preprint", "id": "arxiv:2604.17284", "issued": { "date-parts": [ [ 2026, 4, 19 ] ] }, "note": "First published: 2026-04-19; Recorded arXiv version: v1; Recorded version updated: 2026-04-19T06:55:59Z.", "title": "HalluClear: Diagnosing, Evaluating and Mitigating Hallucinations in GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.17817v1", "archive": "arXiv", "archive_location": "2604.17817", "author": [ { "literal": "Shiquan Zhang" }, { "literal": "Tianyi Zhang" }, { "literal": "Le Fang" }, { "literal": "Simon D'Alfonso" }, { "literal": "Hong Jia" }, { "literal": "Vassilis Kostakos" } ], "citation-key": "arxiv_2604_17817", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "176f4d68e2f5084bee1938ee1399978052cdce2d2250ef06c16ddbf49c977da1", "submitted_at": "2026-04-20T05:15:14Z", "updated_at": "2026-04-20T05:15:14Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.17817v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.17817", "catalog_id": "arxiv:2604.17817", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-20", "identifiers": { "arxiv": "2604.17817" }, "record_updated_at": "2026-04-20" }, "genre": "preprint", "id": "arxiv:2604.17817", "issued": { "date-parts": [ [ 2026, 4, 20 ] ] }, "note": "First published: 2026-04-20; Recorded arXiv version: v1; Recorded version updated: 2026-04-20T05:15:14Z.", "title": "Do LLMs Need to See Everything? A Benchmark and Study of Failures in LLM-driven Smartphone Automation using Screentext vs. Screenshots", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.17849v1", "archive": "arXiv", "archive_location": "2604.17849", "author": [ { "literal": "Gonzalo Gonzalez-Pumariega" }, { "literal": "Saaket Agashe" }, { "literal": "Jiachen Yang" }, { "literal": "Ang Li" }, { "literal": "Xin Eric Wang" } ], "citation-key": "arxiv_2604_17849", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a508db9eba8a7db5b52273bff32a16189f5a8eed19d6563e0f84313188a0aad1", "submitted_at": "2026-04-20T05:59:04Z", "updated_at": "2026-04-20T05:59:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.17849v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.17849", "catalog_id": "arxiv:2604.17849", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-20", "identifiers": { "arxiv": "2604.17849" }, "record_updated_at": "2026-04-20" }, "genre": "preprint", "id": "arxiv:2604.17849", "issued": { "date-parts": [ [ 2026, 4, 20 ] ] }, "note": "First published: 2026-04-20; Recorded arXiv version: v1; Recorded version updated: 2026-04-20T05:59:04Z.", "title": "On the Reliability of Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.18240v1", "archive": "arXiv", "archive_location": "2604.18240", "author": [ { "literal": "Wentao Shi" }, { "literal": "Yu Wang" }, { "literal": "Yuyang Zhao" }, { "literal": "Yuxin Chen" }, { "literal": "Fuli Feng" }, { "literal": "Xueyuan Hao" }, { "literal": "Xi Su" }, { "literal": "Qi Gu" }, { "literal": "Hui Su" }, { "literal": "Xunliang Cai" }, { "literal": "Xiangnan He" } ], "citation-key": "arxiv_2604_18240", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "91a9cdb6c8cb124d2e6033f0decc85e0c73bd972984d973ff687bbdc2be21176", "submitted_at": "2026-04-20T13:23:38Z", "updated_at": "2026-04-20T13:23:38Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.18240v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.18240", "catalog_id": "arxiv:2604.18240", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-20", "identifiers": { "arxiv": "2604.18240" }, "record_updated_at": "2026-04-20" }, "genre": "preprint", "id": "arxiv:2604.18240", "issued": { "date-parts": [ [ 2026, 4, 20 ] ] }, "note": "First published: 2026-04-20; Recorded arXiv version: v1; Recorded version updated: 2026-04-20T13:23:38Z.", "title": "AJ-Bench: Benchmarking Agent-as-a-Judge for Environment-Aware Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.18860v1", "archive": "arXiv", "archive_location": "2604.18860", "author": [ { "literal": "Wenpeng Xu" } ], "citation-key": "arxiv_2604_18860", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8eb499729268ef287a9b9bf73275f526d47dabf2d7de399beaead2fe54cfd74e", "submitted_at": "2026-04-20T21:36:16Z", "updated_at": "2026-04-20T21:36:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.18860v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.18860", "catalog_id": "arxiv:2604.18860", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-20", "identifiers": { "arxiv": "2604.18860" }, "record_updated_at": "2026-04-20" }, "genre": "preprint", "id": "arxiv:2604.18860", "issued": { "date-parts": [ [ 2026, 4, 20 ] ] }, "note": "First published: 2026-04-20; Recorded arXiv version: v1; Recorded version updated: 2026-04-20T21:36:16Z.", "title": "Temporal UI State Inconsistency in Desktop GUI Agents: Formalizing and Defending Against TOCTOU Attacks on Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.19742v2", "archive": "arXiv", "archive_location": "2604.19742", "author": [ { "literal": "Zhiyuan Peng" }, { "literal": "Wei Tao" }, { "literal": "Xin Yin" }, { "literal": "Chenhao Ying" }, { "literal": "Yuan Luo" }, { "literal": "Yiwen Guo" } ], "citation-key": "arxiv_2604_19742", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "74ec7a508a815781cc7e721aa6c5e351f086f6cab46da48e22c2b58f52af8ce0", "submitted_at": "2026-04-21T17:59:16Z", "updated_at": "2026-09-29T10:11:48Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.19742v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.19742", "catalog_id": "arxiv:2604.19742", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-21", "identifiers": { "arxiv": "2604.19742" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2604.19742", "issued": { "date-parts": [ [ 2026, 4, 21 ] ] }, "note": "First published: 2026-04-21; Recorded arXiv version: v2; Recorded version updated: 2026-09-29T10:11:48Z.", "title": "PlayCoder: Making LLM-Generated GUI Code Playable", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.21375v3", "archive": "arXiv", "archive_location": "2604.21375", "author": [ { "literal": "Qijun Han" }, { "literal": "Haoqin Tu" }, { "literal": "Zijun Wang" }, { "literal": "Haoyue Dai" }, { "literal": "Yiyang Zhou" }, { "literal": "Nancy Lau" }, { "literal": "Alvaro A. Cardenas" }, { "literal": "Yuhui Xu" }, { "literal": "Ran Xu" }, { "literal": "Caiming Xiong" }, { "literal": "Zeyu Zheng" }, { "literal": "Huaxiu Yao" }, { "literal": "Yuyin Zhou" }, { "literal": "Cihang Xie" } ], "citation-key": "arxiv_2604_21375", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7e675476121ab3ef51a8627ccfcc86d76eddde73987ea882398f72e3cb31c5d9", "submitted_at": "2026-04-23T07:42:37Z", "updated_at": "2026-09-24T23:58:02Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2604.21375v3" }, "canonical_source_url": "https://arxiv.org/abs/2604.21375", "catalog_id": "arxiv:2604.21375", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-23", "identifiers": { "arxiv": "2604.21375" }, "record_updated_at": "2026-09-24" }, "genre": "preprint", "id": "arxiv:2604.21375", "issued": { "date-parts": [ [ 2026, 4, 23 ] ] }, "note": "First published: 2026-04-23; Recorded arXiv version: v3; Recorded version updated: 2026-09-24T23:58:02Z.", "title": "VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2604.24348v1", "archive": "arXiv", "archive_location": "2604.24348", "author": [ { "literal": "Zheng Wu" }, { "literal": "Yi Hua" }, { "literal": "Zhaoyuan Huang" }, { "literal": "Chenhao Xue" }, { "literal": "Yijie Lu" }, { "literal": "Pengzhou Cheng" }, { "literal": "Zongru Wu" }, { "literal": "Lingzhong Dong" }, { "literal": "Gongshen Liu" }, { "literal": "Xinghao Jiang" }, { "literal": "Zhuosheng Zhang" } ], "citation-key": "arxiv_2604_24348", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "772420683208561df8493c9ed5d68b341e9e098f5729945b972f3555e976d799", "submitted_at": "2026-04-27T11:44:26Z", "updated_at": "2026-04-27T11:44:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.24348v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.24348", "catalog_id": "arxiv:2604.24348", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-27", "identifiers": { "arxiv": "2604.24348" }, "record_updated_at": "2026-04-27" }, "genre": "preprint", "id": "arxiv:2604.24348", "issued": { "date-parts": [ [ 2026, 4, 27 ] ] }, "note": "First published: 2026-04-27; Recorded arXiv version: v1; Recorded version updated: 2026-04-27T11:44:26Z.", "title": "OS-SPEAR: A Toolkit for the Safety, Performance,Efficiency, and Robustness Analysis of OS Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.24441v1", "archive": "arXiv", "archive_location": "2604.24441", "author": [ { "literal": "Hongxin Li" }, { "literal": "Xiping Wang" }, { "literal": "Jingran Su" }, { "literal": "Zheng Ju" }, { "literal": "Yuntao Chen" }, { "literal": "Qing Li" }, { "literal": "Zhaoxiang Zhang" } ], "citation-key": "arxiv_2604_24441", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "65325912365effe8adbc39301a378d7526c60d1185691f3a89cb8acd4aa53c3f", "submitted_at": "2026-04-27T13:06:27Z", "updated_at": "2026-04-27T13:06:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.24441v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.24441", "catalog_id": "arxiv:2604.24441", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-27", "identifiers": { "arxiv": "2604.24441" }, "record_updated_at": "2026-04-27" }, "genre": "preprint", "id": "arxiv:2604.24441", "issued": { "date-parts": [ [ 2026, 4, 27 ] ] }, "note": "First published: 2026-04-27; Recorded arXiv version: v1; Recorded version updated: 2026-04-27T13:06:27Z.", "title": "AutoGUI-v2: A Comprehensive Multi-Modal GUI Functionality Understanding Benchmark", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.24964v1", "archive": "arXiv", "archive_location": "2604.24964", "author": [ { "literal": "Lawrence Keunho Jang" }, { "literal": "Jing Yu Koh" }, { "literal": "Daniel Fried" }, { "literal": "Ruslan Salakhutdinov" } ], "citation-key": "arxiv_2604_24964", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "db70a5736d04d26f10afe3d9a0cfb08955f65f69480a0345ff13b30b2e83df23", "submitted_at": "2026-04-27T20:05:41Z", "updated_at": "2026-04-27T20:05:41Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.24964v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.24964", "catalog_id": "arxiv:2604.24964", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-27", "identifiers": { "arxiv": "2604.24964" }, "record_updated_at": "2026-04-27" }, "genre": "preprint", "id": "arxiv:2604.24964", "issued": { "date-parts": [ [ 2026, 4, 27 ] ] }, "note": "First published: 2026-04-27; Recorded arXiv version: v1; Recorded version updated: 2026-04-27T20:05:41Z.", "title": "Odysseys: Benchmarking Web Agents on Realistic Long Horizon Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.25380v2", "archive": "arXiv", "archive_location": "2604.25380", "author": [ { "literal": "Enqi Liu" }, { "literal": "Liyuan Pan" }, { "literal": "Zhi Gao" }, { "literal": "Yan Yang" }, { "literal": "Chenrui Shi" }, { "literal": "Yang Liu" }, { "literal": "Jingrong Wu" }, { "literal": "Qing Li" } ], "citation-key": "arxiv_2604_25380", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "54b49f54319922268ba880a1c1e8a27ffa7597e228b18be04caa1a6952a6456a", "submitted_at": "2026-04-28T08:43:58Z", "updated_at": "2026-05-08T04:29:20Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2604.25380v2" }, "canonical_source_url": "https://arxiv.org/abs/2604.25380", "catalog_id": "arxiv:2604.25380", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-28", "identifiers": { "arxiv": "2604.25380" }, "record_updated_at": "2026-05-08" }, "genre": "preprint", "id": "arxiv:2604.25380", "issued": { "date-parts": [ [ 2026, 4, 28 ] ] }, "note": "First published: 2026-04-28; Recorded arXiv version: v2; Recorded version updated: 2026-05-08T04:29:20Z.", "title": "Benchmarking and Improving GUI Agents in High-Dynamic Environments", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2604.25562v1", "archive": "arXiv", "archive_location": "2604.25562", "author": [ { "literal": "Mengyao Du" }, { "literal": "Han Fang" }, { "literal": "Haokai Ma" }, { "literal": "Jiahao Chen" }, { "literal": "Kai Xu" }, { "literal": "Quanjun Yin" }, { "literal": "Ee-Chien Chang" } ], "citation-key": "arxiv_2604_25562", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d2610c2658221d7da0bd888bf7b143ca6c4175791058570f6562cfa0c8c3a946", "submitted_at": "2026-04-28T12:32:21Z", "updated_at": "2026-04-28T12:32:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.25562v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.25562", "catalog_id": "arxiv:2604.25562", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-28", "identifiers": { "arxiv": "2604.25562" }, "record_updated_at": "2026-04-28" }, "genre": "preprint", "id": "arxiv:2604.25562", "issued": { "date-parts": [ [ 2026, 4, 28 ] ] }, "note": "First published: 2026-04-28; Recorded arXiv version: v1; Recorded version updated: 2026-04-28T12:32:21Z.", "title": "SnapGuard: Lightweight Prompt Injection Detection for Screenshot-Based Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.26020v1", "archive": "arXiv", "archive_location": "2604.26020", "author": [ { "literal": "Alice Gao" }, { "literal": "Weixi Tong" }, { "literal": "Rishab Vempati" }, { "literal": "Katharina Reinecke" }, { "literal": "R. Benjamin Shapiro" }, { "literal": "Tianyi Zhang" }, { "literal": "Jason Wu" } ], "citation-key": "arxiv_2604_26020", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bb6d24f7dc38921da883c4850b37f1c7eb48a03de920356d4ec2e095dbb44b04", "submitted_at": "2026-04-28T18:04:11Z", "updated_at": "2026-04-28T18:04:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.26020v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.26020", "catalog_id": "arxiv:2604.26020", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-28", "identifiers": { "arxiv": "2604.26020" }, "record_updated_at": "2026-04-28" }, "genre": "preprint", "id": "arxiv:2604.26020", "issued": { "date-parts": [ [ 2026, 4, 28 ] ] }, "note": "First published: 2026-04-28; Recorded arXiv version: v1; Recorded version updated: 2026-04-28T18:04:11Z.", "title": "Training Computer Use Agents to Assess the Usability of Graphical User Interfaces", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.26148v3", "archive": "arXiv", "archive_location": "2604.26148", "author": [ { "literal": "Chen Liang" }, { "literal": "Xirui Jiang" }, { "literal": "Naihao Deng" }, { "literal": "Eytan Adar" }, { "literal": "Anhong Guo" } ], "citation-key": "arxiv_2604_26148", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5b572e37eefab40a6a21fa4e4589324380ad9f66873a4da5746fc0f46aaeb1f6", "submitted_at": "2026-04-28T22:15:06Z", "updated_at": "2026-08-11T15:48:53Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2604.26148v3" }, "canonical_source_url": "https://arxiv.org/abs/2604.26148", "catalog_id": "arxiv:2604.26148", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-28", "identifiers": { "arxiv": "2604.26148" }, "record_updated_at": "2026-08-11" }, "genre": "preprint", "id": "arxiv:2604.26148", "issued": { "date-parts": [ [ 2026, 4, 28 ] ] }, "note": "First published: 2026-04-28; Recorded arXiv version: v3; Recorded version updated: 2026-08-11T15:48:53Z.", "title": "Beyond Screenshots: Evaluating VLMs' Understanding of UI Animations", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2604.27253v1", "archive": "arXiv", "archive_location": "2604.27253", "author": [ { "literal": "Fazle Elahi Faisal" }, { "literal": "Qianhui Wu" }, { "literal": "Baolin Peng" }, { "literal": "Jianfeng Gao" } ], "citation-key": "arxiv_2604_27253", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "276d8a62192c0ad576a1d986bb27d3978963b93466a822858e3313957dd83a81", "submitted_at": "2026-04-29T22:57:35Z", "updated_at": "2026-04-29T22:57:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.27253v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.27253", "catalog_id": "arxiv:2604.27253", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-29", "identifiers": { "arxiv": "2604.27253" }, "record_updated_at": "2026-04-29" }, "genre": "preprint", "id": "arxiv:2604.27253", "issued": { "date-parts": [ [ 2026, 4, 29 ] ] }, "note": "First published: 2026-04-29; Recorded arXiv version: v1; Recorded version updated: 2026-04-29T22:57:35Z.", "title": "AutoSurfer -- Teaching Web Agents through Comprehensive Surfing, Learning, and Modeling", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.27776v1", "archive": "arXiv", "archive_location": "2604.27776", "author": [ { "literal": "Jinchao Li" }, { "literal": "Yunxin Li" }, { "literal": "Chenrui Zhao" }, { "literal": "Zhenran Xu" }, { "literal": "Baotian Hu" }, { "literal": "Min Zhang" } ], "citation-key": "arxiv_2604_27776", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2120860b9c735466a1538ad26c7cda0bc436ab0c6204febdefa08ad66f0036b7", "submitted_at": "2026-04-30T12:13:27Z", "updated_at": "2026-04-30T12:13:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2604.27776v1" }, "canonical_source_url": "https://arxiv.org/abs/2604.27776", "catalog_id": "arxiv:2604.27776", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-30", "identifiers": { "arxiv": "2604.27776" }, "record_updated_at": "2026-04-30" }, "genre": "preprint", "id": "arxiv:2604.27776", "issued": { "date-parts": [ [ 2026, 4, 30 ] ] }, "note": "First published: 2026-04-30; Recorded arXiv version: v1; Recorded version updated: 2026-04-30T12:13:27Z.", "title": "WindowsWorld: A Process-Centric Benchmark of Autonomous GUI Agents in Professional Cross-Application Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2604.27996v3", "archive": "arXiv", "archive_location": "2604.27996", "author": [ { "literal": "Jackson Vonderhorst" }, { "literal": "Kuangshi Ai" }, { "literal": "Haichao Miao" }, { "literal": "Shusen Liu" }, { "literal": "Chaoli Wang" } ], "citation-key": "arxiv_2604_27996", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8e7b07a811df7e74a28f91a5b6c59c69b7ba066bd6c64452a299e678a3187733", "submitted_at": "2026-04-30T15:22:28Z", "updated_at": "2026-06-26T20:31:32Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2604.27996v3" }, "canonical_source_url": "https://arxiv.org/abs/2604.27996", "catalog_id": "arxiv:2604.27996", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-30", "identifiers": { "arxiv": "2604.27996" }, "record_updated_at": "2026-06-26" }, "genre": "preprint", "id": "arxiv:2604.27996", "issued": { "date-parts": [ [ 2026, 4, 30 ] ] }, "note": "First published: 2026-04-30; Recorded arXiv version: v3; Recorded version updated: 2026-06-26T20:31:32Z.", "title": "Exploring LLM Agent Designs and Interaction Modalities for Scientific Visualization", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2605.00551v1", "archive": "arXiv", "archive_location": "2605.00551", "author": [ { "literal": "Michito Takeshita" }, { "literal": "Takuro Kawada" }, { "literal": "Takumi Ohashi" }, { "literal": "Shunsuke Kitada" }, { "literal": "Hitoshi Iyatomi" } ], "citation-key": "arxiv_2605_00551", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "87ef518590e29328ce752791df974c9e2215eac656fcf43fed97dd3e847f3cf7", "submitted_at": "2026-05-01T10:16:00Z", "updated_at": "2026-05-01T10:16:00Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.00551v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.00551", "catalog_id": "arxiv:2605.00551", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-01", "identifiers": { "arxiv": "2605.00551" }, "record_updated_at": "2026-05-01" }, "genre": "preprint", "id": "arxiv:2605.00551", "issued": { "date-parts": [ [ 2026, 5, 1 ] ] }, "note": "First published: 2026-05-01; Recorded arXiv version: v1; Recorded version updated: 2026-05-01T10:16:00Z.", "title": "A11y-Compressor: A Framework for Enhancing the Efficiency of GUI Agent Observations through Visual Context Reconstruction and Redundancy Reduction", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.05509v1", "archive": "arXiv", "archive_location": "2605.05509", "author": [ { "literal": "Sohom Datta" }, { "literal": "Alex Nahapetyan" }, { "literal": "William Enck" }, { "literal": "Alexandros Kapravelos" } ], "citation-key": "arxiv_2605_05509", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0ff44befcfdb97b67d1680c97784f9e9665d8d9879c093984f42d06c9516a342", "submitted_at": "2026-05-06T23:19:38Z", "updated_at": "2026-05-06T23:19:38Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.05509v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.05509", "catalog_id": "arxiv:2605.05509", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-06", "identifiers": { "arxiv": "2605.05509" }, "record_updated_at": "2026-05-06" }, "genre": "preprint", "id": "arxiv:2605.05509", "issued": { "date-parts": [ [ 2026, 5, 6 ] ] }, "note": "First published: 2026-05-06; Recorded arXiv version: v1; Recorded version updated: 2026-05-06T23:19:38Z.", "title": "WAAA! Web Adversaries Against Agentic Browsers", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.09497v1", "archive": "arXiv", "archive_location": "2605.09497", "author": [ { "literal": "Yilin Zhang" }, { "literal": "Yingkai Hua" }, { "literal": "Chunyu Wei" }, { "literal": "Xin Wang" }, { "literal": "Yueguo Chen" } ], "citation-key": "arxiv_2605_09497", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b57d784c70b50d2c25055c6d139c0a5acfc97da3a4effbd63536b72d6df38a69", "submitted_at": "2026-05-10T12:11:52Z", "updated_at": "2026-05-10T12:11:52Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.09497v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.09497", "catalog_id": "arxiv:2605.09497", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-10", "identifiers": { "arxiv": "2605.09497" }, "record_updated_at": "2026-05-10" }, "genre": "preprint", "id": "arxiv:2605.09497", "issued": { "date-parts": [ [ 2026, 5, 10 ] ] }, "note": "First published: 2026-05-10; Recorded arXiv version: v1; Recorded version updated: 2026-05-10T12:11:52Z.", "title": "Don't Click That: Teaching Web Agents to Resist Deceptive Interfaces", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.10347v2", "archive": "arXiv", "archive_location": "2605.10347", "author": [ { "literal": "Weikai Xu" }, { "literal": "Kun Huang" }, { "literal": "Yunren Feng" }, { "literal": "Jiaxing Li" }, { "literal": "Yuhan Chen" }, { "literal": "Yuxuan Liu" }, { "literal": "Zhizheng Jiang" }, { "literal": "Heng Qu" }, { "literal": "Pengzhi Gao" }, { "literal": "Wei Liu" }, { "literal": "Jian Luan" }, { "literal": "Xiaolin Hu" }, { "literal": "Bo An" } ], "citation-key": "arxiv_2605_10347", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4374ec1b7c44804bb78ea56bd0aa3214083419f37780d79c5fcfb68468727bbf", "submitted_at": "2026-05-11T10:49:31Z", "updated_at": "2026-05-22T05:43:30Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2605.10347v2" }, "canonical_source_url": "https://arxiv.org/abs/2605.10347", "catalog_id": "arxiv:2605.10347", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-11", "identifiers": { "arxiv": "2605.10347" }, "record_updated_at": "2026-05-22" }, "genre": "preprint", "id": "arxiv:2605.10347", "issued": { "date-parts": [ [ 2026, 5, 11 ] ] }, "note": "First published: 2026-05-11; Recorded arXiv version: v2; Recorded version updated: 2026-05-22T05:43:30Z.", "title": "How Mobile World Model Guides GUI Agents?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2605.10966v1", "archive": "arXiv", "archive_location": "2605.10966", "author": [ { "literal": "Chiyeong Heo" }, { "literal": "Jaechang Kim" }, { "literal": "Junhyuk Kwon" }, { "literal": "Hoyoung Kim" }, { "literal": "Dongmin Park" }, { "literal": "Jonghyun Lee" }, { "literal": "Jungseul Ok" } ], "citation-key": "arxiv_2605_10966", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0bd75e6bf9d5a8185a2204cdd99d42f9d96c102f7297abcadb9f8425e9246349", "submitted_at": "2026-05-08T10:57:19Z", "updated_at": "2026-05-08T10:57:19Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.10966v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.10966", "catalog_id": "arxiv:2605.10966", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-05-08", "identifiers": { "arxiv": "2605.10966" }, "record_updated_at": "2026-05-08" }, "genre": "preprint", "id": "arxiv:2605.10966", "issued": { "date-parts": [ [ 2026, 5, 8 ] ] }, "note": "First published: 2026-05-08; Recorded arXiv version: v1; Recorded version updated: 2026-05-08T10:57:19Z.", "title": "MMTB: Evaluating Terminal Agents on Multimedia-File Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.12493v1", "archive": "arXiv", "archive_location": "2605.12493", "author": [ { "literal": "Di Wu" }, { "literal": "Zixiang Ji" }, { "literal": "Asmi Kawatkar" }, { "literal": "Bryan Kwan" }, { "literal": "Jia-Chen Gu" }, { "literal": "Nanyun Peng" }, { "literal": "Kai-Wei Chang" } ], "citation-key": "arxiv_2605_12493", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5893d1d9205776667a1e1d89720ec3baf55fafe80cc9c6201ca97088f44a10ac", "submitted_at": "2026-05-12T17:59:34Z", "updated_at": "2026-05-12T17:59:34Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.12493v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.12493", "catalog_id": "arxiv:2605.12493", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-12", "identifiers": { "arxiv": "2605.12493" }, "record_updated_at": "2026-05-12" }, "genre": "preprint", "id": "arxiv:2605.12493", "issued": { "date-parts": [ [ 2026, 5, 12 ] ] }, "note": "First published: 2026-05-12; Recorded arXiv version: v1; Recorded version updated: 2026-05-12T17:59:34Z.", "title": "LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.12501v1", "archive": "arXiv", "archive_location": "2605.12501", "author": [ { "literal": "Miaosen Zhang" }, { "literal": "Xiaohan Zhao" }, { "literal": "Zhihong Tan" }, { "literal": "Zhou Huoshen" }, { "literal": "Yijia Fan" }, { "literal": "Yifan Yang" }, { "literal": "Kai Qiu" }, { "literal": "Bei Liu" }, { "literal": "Justin Wagle" }, { "literal": "Chenzhong Yin" }, { "literal": "Mingxi Cheng" }, { "literal": "Ji Li" }, { "literal": "Qi Dai" }, { "literal": "Chong Luo" }, { "literal": "Xu Yang" }, { "literal": "Xin Geng" }, { "literal": "Baining Guo" } ], "citation-key": "arxiv_2605_12501", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1fb57ba3cd858513d1da58d64a7c85388f12e331c9ea9f89204be5f43e6fec38", "submitted_at": "2026-05-12T17:59:58Z", "updated_at": "2026-05-12T17:59:58Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.12501v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.12501", "catalog_id": "arxiv:2605.12501", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-12", "identifiers": { "arxiv": "2605.12501" }, "record_updated_at": "2026-05-12" }, "genre": "preprint", "id": "arxiv:2605.12501", "issued": { "date-parts": [ [ 2026, 5, 12 ] ] }, "note": "First published: 2026-05-12; Recorded arXiv version: v1; Recorded version updated: 2026-05-12T17:59:58Z.", "title": "Covering Human Action Space for Computer Use: Data Synthesis and Benchmark", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.12549v1", "archive": "arXiv", "archive_location": "2605.12549", "author": [ { "literal": "Jiaping Lin" }, { "literal": "Fei Shen" }, { "literal": "Junzhe Li" }, { "literal": "Ping Nie" }, { "literal": "Fei Yu" }, { "literal": "Ming Li" }, { "literal": "Haizhou Li" } ], "citation-key": "arxiv_2605_12549", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cb60b76fca3a8825030b1a31a7b97ff31e06f30ac3852f243efe440f965e5578", "submitted_at": "2026-05-10T07:04:07Z", "updated_at": "2026-05-10T07:04:07Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.12549v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.12549", "catalog_id": "arxiv:2605.12549", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-10", "identifiers": { "arxiv": "2605.12549" }, "record_updated_at": "2026-05-10" }, "genre": "preprint", "id": "arxiv:2605.12549", "issued": { "date-parts": [ [ 2026, 5, 10 ] ] }, "note": "First published: 2026-05-10; Recorded arXiv version: v1; Recorded version updated: 2026-05-10T07:04:07Z.", "title": "What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.15777v2", "archive": "arXiv", "archive_location": "2605.15777", "author": [ { "literal": "Kean Shi" }, { "literal": "Zihang Li" }, { "literal": "Tianyi Ma" }, { "literal": "Zengji Tu" }, { "literal": "Jialong Wu" }, { "literal": "Xinbo Xu" }, { "literal": "Qingyao Yang" }, { "literal": "Ruoyu Wu" }, { "literal": "Weichu Xie" }, { "literal": "Ming Wu" }, { "literal": "Jason Zeng" }, { "literal": "Michael Heinrich" }, { "literal": "Elvis Zhang" }, { "literal": "Liang Chen" }, { "literal": "Kuan Li" }, { "literal": "Baobao Chang" } ], "citation-key": "arxiv_2605_15777", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c4ee2439dd31a375b6f64cbc7f1d0ff75c87e1346b1627a53db145ae42682f8a", "submitted_at": "2026-05-15T09:35:15Z", "updated_at": "2026-05-24T16:35:56Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2605.15777v2" }, "canonical_source_url": "https://arxiv.org/abs/2605.15777", "catalog_id": "arxiv:2605.15777", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-15", "identifiers": { "arxiv": "2605.15777" }, "record_updated_at": "2026-05-24" }, "genre": "preprint", "id": "arxiv:2605.15777", "issued": { "date-parts": [ [ 2026, 5, 15 ] ] }, "note": "First published: 2026-05-15; Recorded arXiv version: v2; Recorded version updated: 2026-05-24T16:35:56Z.", "title": "SaaS-Bench: Can Computer-Use Agents Leverage Real-World SaaS to Solve Professional Workflows?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2605.16402v1", "archive": "arXiv", "archive_location": "2605.16402", "author": [ { "literal": "Haoren Zhao" }, { "literal": "Tianyi Chen" }, { "literal": "Zhen Wang" } ], "citation-key": "arxiv_2605_16402", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "671f33dd87a7ebf9233494ef6fa0efd9c9772a3951f3c218bf4de822825f7af2", "submitted_at": "2026-05-13T02:48:52Z", "updated_at": "2026-05-13T02:48:52Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.16402v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.16402", "catalog_id": "arxiv:2605.16402", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-13", "identifiers": { "arxiv": "2605.16402" }, "record_updated_at": "2026-05-13" }, "genre": "preprint", "id": "arxiv:2605.16402", "issued": { "date-parts": [ [ 2026, 5, 13 ] ] }, "note": "First published: 2026-05-13; Recorded arXiv version: v1; Recorded version updated: 2026-05-13T02:48:52Z.", "title": "WinDeskGround: A Benchmark for Robust GUI Grounding in Complex Multi-Window Desktop Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.16883v1", "archive": "arXiv", "archive_location": "2605.16883", "author": [ { "literal": "Shilong Jin" }, { "literal": "Lanjun Wang" }, { "literal": "Zhuosheng Zhang" } ], "citation-key": "arxiv_2605_16883", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "52c1bf4c614ac05a55f2c2c66d3ac4eeb1bf31743814a45eb9f0ce064b6baaab", "submitted_at": "2026-05-16T08:51:57Z", "updated_at": "2026-05-16T08:51:57Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.16883v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.16883", "catalog_id": "arxiv:2605.16883", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-16", "identifiers": { "arxiv": "2605.16883" }, "record_updated_at": "2026-05-16" }, "genre": "preprint", "id": "arxiv:2605.16883", "issued": { "date-parts": [ [ 2026, 5, 16 ] ] }, "note": "First published: 2026-05-16; Recorded arXiv version: v1; Recorded version updated: 2026-05-16T08:51:57Z.", "title": "SE-GA: Memory-Augmented Self-Evolution for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.17656v1", "archive": "arXiv", "archive_location": "2605.17656", "author": [ { "literal": "Athar Parvez" }, { "literal": "Muhammad Jawad Mufti" }, { "literal": "Muqaddas Gull" }, { "literal": "Omar Hammad" } ], "citation-key": "arxiv_2605_17656", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ff95e0b071508592190ba236741c1c01200dd38fdc74c31b2cc4d1bf8143ae9c", "submitted_at": "2026-05-17T21:14:32Z", "updated_at": "2026-05-17T21:14:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.17656v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.17656", "catalog_id": "arxiv:2605.17656", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-17", "identifiers": { "arxiv": "2605.17656" }, "record_updated_at": "2026-05-17" }, "genre": "preprint", "id": "arxiv:2605.17656", "issued": { "date-parts": [ [ 2026, 5, 17 ] ] }, "note": "First published: 2026-05-17; Recorded arXiv version: v1; Recorded version updated: 2026-05-17T21:14:32Z.", "title": "MUIAnno: An Expert-Annotated Dataset and Evaluation Benchmark for Mobile UI Understanding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.18048v1", "archive": "arXiv", "archive_location": "2605.18048", "author": [ { "literal": "Jingjing Liu" }, { "literal": "Ziye Huang" }, { "literal": "Zihao Cheng" }, { "literal": "Zeming Liu" }, { "literal": "Jiahong Wu" }, { "literal": "Yuhang Guo" }, { "literal": "Kehai Chen" }, { "literal": "Yunhong Wang" }, { "literal": "Haifeng Wang" } ], "citation-key": "arxiv_2605_18048", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "017909a697c6c661d3486d960042f4d691f70a660a60083e70512de34690cb51", "submitted_at": "2026-05-18T08:36:04Z", "updated_at": "2026-05-18T08:36:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.18048v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.18048", "catalog_id": "arxiv:2605.18048", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-18", "identifiers": { "arxiv": "2605.18048" }, "record_updated_at": "2026-05-18" }, "genre": "preprint", "id": "arxiv:2605.18048", "issued": { "date-parts": [ [ 2026, 5, 18 ] ] }, "note": "First published: 2026-05-18; Recorded arXiv version: v1; Recorded version updated: 2026-05-18T08:36:04Z.", "title": "DocOS: Towards Proactive Document-Guided Actions in GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.18652v1", "archive": "arXiv", "archive_location": "2605.18652", "author": [ { "literal": "Ziyun Zeng" }, { "literal": "Hang Hua" }, { "literal": "Bocheng Zou" }, { "literal": "Mu Cai" }, { "literal": "Rogerio Feris" }, { "literal": "Jiebo Luo" } ], "citation-key": "arxiv_2605_18652", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d96c8844274f79adea167169a0853e3a7cd82fee4a8a666913f97898a0836e7c", "submitted_at": "2026-05-18T16:57:36Z", "updated_at": "2026-05-18T16:57:36Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.18652v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.18652", "catalog_id": "arxiv:2605.18652", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-18", "identifiers": { "arxiv": "2605.18652" }, "record_updated_at": "2026-05-18" }, "genre": "preprint", "id": "arxiv:2605.18652", "issued": { "date-parts": [ [ 2026, 5, 18 ] ] }, "note": "First published: 2026-05-18; Recorded arXiv version: v1; Recorded version updated: 2026-05-18T16:57:36Z.", "title": "MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.18758v2", "archive": "arXiv", "archive_location": "2605.18758", "author": [ { "literal": "Felix Henry" }, { "literal": "Lihan Lin" }, { "literal": "Jiangyou Zhu" }, { "literal": "Yangfan" }, { "literal": "Bingqian Zhang" }, { "literal": "Min Chen" }, { "literal": "Shiyu Huang" } ], "citation-key": "arxiv_2605_18758", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "99e8e0039ec99ef8f6a6ef3900cb74a62ddd5ae4f8df30629110969290028e6a", "submitted_at": "2026-04-03T08:57:06Z", "updated_at": "2026-09-28T02:42:06Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2605.18758v2" }, "canonical_source_url": "https://arxiv.org/abs/2605.18758", "catalog_id": "arxiv:2605.18758", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-04-03", "identifiers": { "arxiv": "2605.18758" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2605.18758", "issued": { "date-parts": [ [ 2026, 4, 3 ] ] }, "note": "First published: 2026-04-03; Recorded arXiv version: v2; Recorded version updated: 2026-09-28T02:42:06Z.", "title": "OmniGUI: Benchmarking GUI Agents in Omni-Modal Smartphone Environments", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2605.19769v1", "archive": "arXiv", "archive_location": "2605.19769", "author": [ { "literal": "Jinbiao Wei" }, { "literal": "Qianran Ma" }, { "literal": "Yilun Zhao" }, { "literal": "Xiao Zhou" }, { "literal": "Kangqi Ni" }, { "literal": "Guo Gan" }, { "literal": "Arman Cohan" } ], "citation-key": "arxiv_2605_19769", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0a93d880793018d1b47a39e0487e3be76550fd2b49de9535b2913e9f08159074", "submitted_at": "2026-05-19T12:40:29Z", "updated_at": "2026-05-19T12:40:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.19769v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.19769", "catalog_id": "arxiv:2605.19769", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-19", "identifiers": { "arxiv": "2605.19769" }, "record_updated_at": "2026-05-19" }, "genre": "preprint", "id": "arxiv:2605.19769", "issued": { "date-parts": [ [ 2026, 5, 19 ] ] }, "note": "First published: 2026-05-19; Recorded arXiv version: v1; Recorded version updated: 2026-05-19T12:40:29Z.", "title": "OpenComputer: Verifiable Software Worlds for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.25160v2", "archive": "arXiv", "archive_location": "2605.25160", "author": [ { "literal": "Guohong Liu" }, { "literal": "Jialei Ye" }, { "literal": "Pengzhi Gao" }, { "literal": "Wei Liu" }, { "literal": "Jian Luan" }, { "literal": "Yunxin Liu" }, { "literal": "Yuanchun Li" } ], "citation-key": "arxiv_2605_25160", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "eb40c8bf005d56139d263e1b5dc4553ddcd68522124686b7f0865e6d8d1725ad", "submitted_at": "2026-05-24T16:33:14Z", "updated_at": "2026-06-18T02:24:07Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2605.25160v2" }, "canonical_source_url": "https://arxiv.org/abs/2605.25160", "catalog_id": "arxiv:2605.25160", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-24", "identifiers": { "arxiv": "2605.25160" }, "record_updated_at": "2026-06-18" }, "genre": "preprint", "id": "arxiv:2605.25160", "issued": { "date-parts": [ [ 2026, 5, 24 ] ] }, "note": "First published: 2026-05-24; Recorded arXiv version: v2; Recorded version updated: 2026-06-18T02:24:07Z.", "title": "ScaleWoB: Guiding GUI Agents with Coding Agents via Large-Scale Environmental Synthesis", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2605.26086v1", "archive": "arXiv", "archive_location": "2605.26086", "author": [ { "literal": "Yusong Lin" }, { "literal": "Xinyuan Liang" }, { "literal": "Haiyang Wang" }, { "literal": "Qipeng Gu" }, { "literal": "Siqi Cheng" }, { "literal": "Jiangui Chen" }, { "literal": "Shuzhe Wu" }, { "literal": "Feiyang Pan" }, { "literal": "Lue Fan" }, { "literal": "Sanyuan Zhao" }, { "literal": "Dandan Tu" } ], "citation-key": "arxiv_2605_26086", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "27999553928305cfa4223011feb3f9c2decda8a861e5800865537a81c7c08832", "submitted_at": "2026-05-25T17:50:04Z", "updated_at": "2026-05-25T17:50:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.26086v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.26086", "catalog_id": "arxiv:2605.26086", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-25", "identifiers": { "arxiv": "2605.26086" }, "record_updated_at": "2026-05-25" }, "genre": "preprint", "id": "arxiv:2605.26086", "issued": { "date-parts": [ [ 2026, 5, 25 ] ] }, "note": "First published: 2026-05-25; Recorded arXiv version: v1; Recorded version updated: 2026-05-25T17:50:04Z.", "title": "Claw-Anything: Benchmarking Always-On Personal Assistants with Broader Access to User's Digital World", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.26114v3", "archive": "arXiv", "archive_location": "2605.26114", "author": [ { "literal": "Dingbang Wu" }, { "literal": "Rui Hao" }, { "literal": "Haiyang Wang" }, { "literal": "Shuzhe Wu" }, { "literal": "Han Xiao" }, { "literal": "Zhenghong Li" }, { "literal": "Bojiang Zhou" }, { "literal": "Zheng Ju" }, { "literal": "Zichen Liu" }, { "literal": "Lue Fan" }, { "literal": "Zhaoxiang Zhang" } ], "citation-key": "arxiv_2605_26114", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "145c7012fb15a85b203924ecd559c8b20d18b51bf79805a6751fec75d6f68dec", "submitted_at": "2026-05-25T17:59:49Z", "updated_at": "2026-09-23T17:59:57Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2605.26114v3" }, "canonical_source_url": "https://arxiv.org/abs/2605.26114", "catalog_id": "arxiv:2605.26114", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-25", "identifiers": { "arxiv": "2605.26114" }, "record_updated_at": "2026-09-23" }, "genre": "preprint", "id": "arxiv:2605.26114", "issued": { "date-parts": [ [ 2026, 5, 25 ] ] }, "note": "First published: 2026-05-25; Recorded arXiv version: v3; Recorded version updated: 2026-09-23T17:59:57Z.", "title": "MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2605.26144v4", "archive": "arXiv", "archive_location": "2605.26144", "author": [ { "literal": "JunJia Guo" }, { "literal": "Yuhang Yao" }, { "literal": "Chang Li" }, { "literal": "Zimeng Pan" }, { "literal": "Jiawei" }, { "literal": "Zhou" }, { "literal": "Jingdi Chen" } ], "citation-key": "arxiv_2605_26144", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "64f8e09fc62f080a75aec1928b82187063cfc3ad8467290608347622410b6460", "submitted_at": "2026-05-22T20:29:12Z", "updated_at": "2026-09-29T00:12:17Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2605.26144v4" }, "canonical_source_url": "https://arxiv.org/abs/2605.26144", "catalog_id": "arxiv:2605.26144", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-05-22", "identifiers": { "arxiv": "2605.26144" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2605.26144", "issued": { "date-parts": [ [ 2026, 5, 22 ] ] }, "note": "First published: 2026-05-22; Recorded arXiv version: v4; Recorded version updated: 2026-09-29T00:12:17Z.", "title": "VISTA: An End-to-End Benchmark for Visual Spec-to-Web-App Coding Agents", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2605.27761v1", "archive": "arXiv", "archive_location": "2605.27761", "author": [ { "literal": "Yifan Sui" }, { "literal": "Xin Huang" }, { "literal": "Hongbing Li" }, { "literal": "Fang Xu" }, { "literal": "Jiahe Lv" }, { "literal": "Haolong Yan" }, { "literal": "Yeqing Shen" }, { "literal": "Litao Liu" }, { "literal": "Zhimin Fan" }, { "literal": "Ziyang Meng" }, { "literal": "Jia Wang" }, { "literal": "Junbo Qi" }, { "literal": "Kaijun Tan" }, { "literal": "Zheng Ge" }, { "literal": "Xiangyu Zhang" }, { "literal": "Daxin Jiang" }, { "literal": "Osamu Yoshie" } ], "citation-key": "arxiv_2605_27761", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0c89b8b5ef5aa17fe747201285104544648449804cc2229a77577d217763f497", "submitted_at": "2026-05-26T23:19:42Z", "updated_at": "2026-05-26T23:19:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.27761v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.27761", "catalog_id": "arxiv:2605.27761", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-26", "identifiers": { "arxiv": "2605.27761" }, "record_updated_at": "2026-05-26" }, "genre": "preprint", "id": "arxiv:2605.27761", "issued": { "date-parts": [ [ 2026, 5, 26 ] ] }, "note": "First published: 2026-05-26; Recorded arXiv version: v1; Recorded version updated: 2026-05-26T23:19:42Z.", "title": "AndroidDaily: A Verifiable Benchmark for Mobile GUI Agents on Real-World Closed-Source Applications", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.28116v1", "archive": "arXiv", "archive_location": "2605.28116", "author": [ { "literal": "Ruoqi Guo" }, { "literal": "Yi Liu" }, { "literal": "Gelei Deng" }, { "literal": "Yiheng Xiong" }, { "literal": "Yuekang Li" }, { "literal": "Ying Zhang" }, { "literal": "Leo Yu Zhang" }, { "literal": "Lida Zhao" }, { "literal": "Ji Jie" }, { "literal": "Yuxiao Lu" } ], "citation-key": "arxiv_2605_28116", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4f39433695b9574b275fc844ca6b10a43baf66aed57048bf841f41bf08d4ac9a", "submitted_at": "2026-05-27T08:09:53Z", "updated_at": "2026-05-27T08:09:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2605.28116v1" }, "canonical_source_url": "https://arxiv.org/abs/2605.28116", "catalog_id": "arxiv:2605.28116", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-27", "identifiers": { "arxiv": "2605.28116" }, "record_updated_at": "2026-05-27" }, "genre": "preprint", "id": "arxiv:2605.28116", "issued": { "date-parts": [ [ 2026, 5, 27 ] ] }, "note": "First published: 2026-05-27; Recorded arXiv version: v1; Recorded version updated: 2026-05-27T08:09:53Z.", "title": "MIRAGE: Context-Aware Prompt Injection against Mobile GUI Agents via User-Generated Content", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2605.28258v2", "archive": "arXiv", "archive_location": "2605.28258", "author": [ { "literal": "Yixu Huang" }, { "literal": "Bo Li" }, { "literal": "Na Li" }, { "literal": "Zhe Wang" }, { "literal": "Kaijie Chen" }, { "literal": "Haonan Ge" }, { "literal": "Qingyi Si" }, { "literal": "Yuanzhe Shen" }, { "literal": "Ruihan Yang" }, { "literal": "Guangjing Wang" }, { "literal": "Hongcheng Guo" } ], "citation-key": "arxiv_2605_28258", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4e03e0ef06c6610eaefeec08b7891a1ab4d5073c0793650d4845735259978723", "submitted_at": "2026-05-27T10:08:48Z", "updated_at": "2026-10-02T05:26:34Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2605.28258v2" }, "canonical_source_url": "https://arxiv.org/abs/2605.28258", "catalog_id": "arxiv:2605.28258", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-05-27", "identifiers": { "arxiv": "2605.28258" }, "record_updated_at": "2026-10-02" }, "genre": "preprint", "id": "arxiv:2605.28258", "issued": { "date-parts": [ [ 2026, 5, 27 ] ] }, "note": "First published: 2026-05-27; Recorded arXiv version: v2; Recorded version updated: 2026-10-02T05:26:34Z.", "title": "GUI Agents for Continual Game Generation", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2605.28646v2", "archive": "arXiv", "archive_location": "2605.28646", "author": [ { "literal": "Yanqiu Zhao" }, { "literal": "Dongying Zheng" }, { "literal": "Kaibo Huang" }, { "literal": "Yukun Wei" }, { "literal": "Zhongliang Yang" }, { "literal": "Linna Zhou" } ], "citation-key": "arxiv_2605_28646", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6eb9b67f7e51edecfbe57af47b6bad445c05129c11c377d32ea84369b430f895", "submitted_at": "2026-05-27T15:51:22Z", "updated_at": "2026-05-29T04:03:46Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2605.28646v2" }, "canonical_source_url": "https://arxiv.org/abs/2605.28646", "catalog_id": "arxiv:2605.28646", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-27", "identifiers": { "arxiv": "2605.28646" }, "record_updated_at": "2026-05-29" }, "genre": "preprint", "id": "arxiv:2605.28646", "issued": { "date-parts": [ [ 2026, 5, 27 ] ] }, "note": "First published: 2026-05-27; Recorded arXiv version: v2; Recorded version updated: 2026-05-29T04:03:46Z.", "title": "MaskClaw: Edge-Side Personalized Privacy Arbitration for GUI Agents with Behavior-Driven Skill Evolution", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2605.29486v3", "archive": "arXiv", "archive_location": "2605.29486", "author": [ { "literal": "Yuxuan Liu" }, { "literal": "Xin Lai" }, { "literal": "Junyi Li" }, { "literal": "Pengyuan Lyu" }, { "literal": "Jason" }, { "literal": "Yiduo Guo" }, { "literal": "Zhengyao Fang" }, { "literal": "Yang Ding" }, { "literal": "Yi Zhang" }, { "literal": "Weinong Wang" }, { "literal": "Huawen Shen" }, { "literal": "Xingran Zhou" }, { "literal": "Liang Wu" }, { "literal": "Fei Tang" }, { "literal": "Sunqi Fan" }, { "literal": "Shangpin Peng" }, { "literal": "Zheng Ruan" }, { "literal": "Anran Zhang" }, { "literal": "Chengquan Zhang" }, { "literal": "Han Hu" }, { "literal": "Benyou Wang" }, { "literal": "Ji-Rong Wen" }, { "literal": "Rui Yan" }, { "literal": "Zhengyang Tang" } ], "citation-key": "arxiv_2605_29486", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a41cb1a93e6aa6ba8fbc7dca55415577ab5880a6592e9b0bee5b12b5a7732cbf", "submitted_at": "2026-05-28T07:14:15Z", "updated_at": "2026-09-28T15:27:12Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2605.29486v3" }, "canonical_source_url": "https://arxiv.org/abs/2605.29486", "catalog_id": "arxiv:2605.29486", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-28", "identifiers": { "arxiv": "2605.29486" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2605.29486", "issued": { "date-parts": [ [ 2026, 5, 28 ] ] }, "note": "First published: 2026-05-28; Recorded arXiv version: v3; Recorded version updated: 2026-09-28T15:27:12Z.", "title": "PhoneWorld: From Real-App Trajectories to Dynamic and Verifiable Environments for Phone-Use Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2606.03103v1", "archive": "arXiv", "archive_location": "2606.03103", "author": [ { "literal": "Wenkai Wang" }, { "literal": "Tao Xiong" }, { "literal": "Jingchen Ni" }, { "literal": "Yunpeng Bao" }, { "literal": "Xiyun Li" }, { "literal": "Tianqi Liu" }, { "literal": "Hongcan Guo" }, { "literal": "Zilong Huang" }, { "literal": "Shengyu Zhang" } ], "citation-key": "arxiv_2606_03103", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bf0fcafeb2d40e489ec80784a3541114b02c6a9077a330265d5c49775eca4582", "submitted_at": "2026-06-02T03:42:34Z", "updated_at": "2026-06-02T03:42:34Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.03103v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.03103", "catalog_id": "arxiv:2606.03103", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-02", "identifiers": { "arxiv": "2606.03103" }, "record_updated_at": "2026-06-02" }, "genre": "preprint", "id": "arxiv:2606.03103", "issued": { "date-parts": [ [ 2026, 6, 2 ] ] }, "note": "First published: 2026-06-02; Recorded arXiv version: v1; Recorded version updated: 2026-06-02T03:42:34Z.", "title": "DeskCraft: Benchmarking Desktop Agents on Professional Workflows and Human-in-the-Loop Collaboration", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.03203v1", "archive": "arXiv", "archive_location": "2606.03203", "author": [ { "literal": "Jia Yu" }, { "literal": "Zilong Wang" }, { "literal": "Xinyang Jiang" }, { "literal": "Dongsheng Li" }, { "literal": "Shuo Wang" } ], "citation-key": "arxiv_2606_03203", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b23c12cafdd2d4a3ef63027c914ed3894c61c555edd89468c56ae40c3ec6cea9", "submitted_at": "2026-06-02T06:02:35Z", "updated_at": "2026-06-02T06:02:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.03203v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.03203", "catalog_id": "arxiv:2606.03203", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-02", "identifiers": { "arxiv": "2606.03203" }, "record_updated_at": "2026-06-02" }, "genre": "preprint", "id": "arxiv:2606.03203", "issued": { "date-parts": [ [ 2026, 6, 2 ] ] }, "note": "First published: 2026-06-02; Recorded arXiv version: v1; Recorded version updated: 2026-06-02T06:02:35Z.", "title": "MedCUA-Bench: A Screenshot-Only Benchmark for Clinical Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.04701v1", "archive": "arXiv", "archive_location": "2606.04701", "author": [ { "literal": "Jiashu Yao" }, { "literal": "Heyan Huang" }, { "literal": "Daiqing Wu" }, { "literal": "Wangke Chen" }, { "literal": "Huaxi Ai" }, { "literal": "Haoyu Wen" }, { "literal": "Zeming Liu" }, { "literal": "Yuhang Guo" } ], "citation-key": "arxiv_2606_04701", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4d5a90333c4f6be0b8d203a2f5938f7e7eefcb09fbe7ea47b85eba8c7c1c422a", "submitted_at": "2026-06-03T10:25:46Z", "updated_at": "2026-06-03T10:25:46Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.04701v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.04701", "catalog_id": "arxiv:2606.04701", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-03", "identifiers": { "arxiv": "2606.04701" }, "record_updated_at": "2026-06-03" }, "genre": "preprint", "id": "arxiv:2606.04701", "issued": { "date-parts": [ [ 2026, 6, 3 ] ] }, "note": "First published: 2026-06-03; Recorded arXiv version: v1; Recorded version updated: 2026-06-03T10:25:46Z.", "title": "Benchmarking Living-Screen-Native GUI Agents on Short-Video Platforms", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.05233v1", "archive": "arXiv", "archive_location": "2606.05233", "author": [ { "literal": "Nicholas Saban" } ], "citation-key": "arxiv_2606_05233", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7b649c0cef63bc15558aa4498a65043644c109faaf6403b610c2132fd3f5706d", "submitted_at": "2026-06-03T01:21:59Z", "updated_at": "2026-06-03T01:21:59Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.05233v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.05233", "catalog_id": "arxiv:2606.05233", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-03", "identifiers": { "arxiv": "2606.05233" }, "record_updated_at": "2026-06-03" }, "genre": "preprint", "id": "arxiv:2606.05233", "issued": { "date-parts": [ [ 2026, 6, 3 ] ] }, "note": "First published: 2026-06-03; Recorded arXiv version: v1; Recorded version updated: 2026-06-03T01:21:59Z.", "title": "Domain-Conditioned Safety in Frontier Computer-Using Agents: A 793-Episode Browser Benchmark, a Coding-Domain Cross-Reference, and a Reproducibility Audit of Recent Red-Teaming", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.06322v1", "archive": "arXiv", "archive_location": "2606.06322", "author": [ { "literal": "Nathan Bout" }, { "literal": "Maxime Langevin" }, { "literal": "Ronan Riochet" } ], "citation-key": "arxiv_2606_06322", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "27ffcb10d966946de9fa2353c4b96798e5f090bdfdc5610f5e8a5c6233cf70ea", "submitted_at": "2026-06-04T15:57:29Z", "updated_at": "2026-06-04T15:57:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.06322v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.06322", "catalog_id": "arxiv:2606.06322", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-04", "identifiers": { "arxiv": "2606.06322" }, "record_updated_at": "2026-06-04" }, "genre": "preprint", "id": "arxiv:2606.06322", "issued": { "date-parts": [ [ 2026, 6, 4 ] ] }, "note": "First published: 2026-06-04; Recorded arXiv version: v1; Recorded version updated: 2026-06-04T15:57:29Z.", "title": "DragOn: A Benchmark and Dataset for Drag-Based GUI Interactions", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.06560v1", "archive": "arXiv", "archive_location": "2606.06560", "author": [ { "literal": "Victor Muryn" }, { "literal": "Maksym Shamrai" }, { "literal": "Sofiia Mazepa" }, { "literal": "Yehor Khodysko" } ], "citation-key": "arxiv_2606_06560", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fd66adb4e99df2348f59e238d219d48cb6e758e694839367d2b9a64675429d87", "submitted_at": "2026-06-04T14:01:32Z", "updated_at": "2026-06-04T14:01:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.06560v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.06560", "catalog_id": "arxiv:2606.06560", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-04", "identifiers": { "arxiv": "2606.06560" }, "record_updated_at": "2026-06-04" }, "genre": "preprint", "id": "arxiv:2606.06560", "issued": { "date-parts": [ [ 2026, 6, 4 ] ] }, "note": "First published: 2026-06-04; Recorded arXiv version: v1; Recorded version updated: 2026-06-04T14:01:32Z.", "title": "MacArena: Benchmarking Computer Use Agents on an Online macOS Environment", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.07027v2", "archive": "arXiv", "archive_location": "2606.07027", "author": [ { "literal": "Haojie Hao" }, { "literal": "Longkun Hao" }, { "literal": "Yihang Lou" }, { "literal": "Yan Bai" }, { "literal": "Zhenyang Li" }, { "literal": "Zhichao Yang" }, { "literal": "Dongshuo Huang" }, { "literal": "Hongyu Lin" }, { "literal": "Lanqing Hong" }, { "literal": "Jiakai Wang" }, { "literal": "Xianglong Liu" } ], "citation-key": "arxiv_2606_07027", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2d530152b210a69cb458fecb036c10621c72daccfbcac7c8306c7e248c87cf14", "submitted_at": "2026-06-05T08:17:28Z", "updated_at": "2026-06-12T07:28:19Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.07027v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.07027", "catalog_id": "arxiv:2606.07027", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-05", "identifiers": { "arxiv": "2606.07027" }, "record_updated_at": "2026-06-12" }, "genre": "preprint", "id": "arxiv:2606.07027", "issued": { "date-parts": [ [ 2026, 6, 5 ] ] }, "note": "First published: 2026-06-05; Recorded arXiv version: v2; Recorded version updated: 2026-06-12T07:28:19Z.", "title": "StainFlow: Entity-Stain Tracking and Evidence Linking for Process Rewards in GUI Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.09426v3", "archive": "arXiv", "archive_location": "2606.09426", "author": [ { "literal": "Wanli Li" }, { "literal": "Bowen Zhou" }, { "literal": "Yunyao Yu" }, { "literal": "Zhou Xu" }, { "literal": "Yifan Yang" }, { "literal": "Dongsheng Li" }, { "literal": "Caihua Shan" } ], "citation-key": "arxiv_2606_09426", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "73b751efd63d98fc81166a1a80cb0c162d68bf1ec4ecca0a1a1a7717ad62e680", "submitted_at": "2026-06-08T12:39:23Z", "updated_at": "2026-07-06T14:25:02Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2606.09426v3" }, "canonical_source_url": "https://arxiv.org/abs/2606.09426", "catalog_id": "arxiv:2606.09426", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-08", "identifiers": { "arxiv": "2606.09426" }, "record_updated_at": "2026-07-06" }, "genre": "preprint", "id": "arxiv:2606.09426", "issued": { "date-parts": [ [ 2026, 6, 8 ] ] }, "note": "First published: 2026-06-08; Recorded arXiv version: v3; Recorded version updated: 2026-07-06T14:25:02Z.", "title": "WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2606.09447v1", "archive": "arXiv", "archive_location": "2606.09447", "author": [ { "literal": "Bojie Rong" }, { "literal": "Zheyu Shen" }, { "literal": "Qiaoping Wang" }, { "literal": "Pengfei Kang" }, { "literal": "Yang Xu" }, { "literal": "Yawen Wei" }, { "literal": "Hanyu Wu" }, { "literal": "Zhi Zhao" }, { "literal": "Leihao Pei" }, { "literal": "Linquan Jiang" } ], "citation-key": "arxiv_2606_09447", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0a852c77650fa54f234794c6cd27777f4e7219b8389b509d807e27db16138a1b", "submitted_at": "2026-06-08T12:55:42Z", "updated_at": "2026-06-08T12:55:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.09447v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.09447", "catalog_id": "arxiv:2606.09447", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-08", "identifiers": { "arxiv": "2606.09447" }, "record_updated_at": "2026-06-08" }, "genre": "preprint", "id": "arxiv:2606.09447", "issued": { "date-parts": [ [ 2026, 6, 8 ] ] }, "note": "First published: 2026-06-08; Recorded arXiv version: v1; Recorded version updated: 2026-06-08T12:55:42Z.", "title": "AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.09764v1", "archive": "arXiv", "archive_location": "2606.09764", "author": [ { "literal": "Lawrence Keunho Jang" }, { "literal": "Mareks Woodside" }, { "literal": "Geronimo Carom" }, { "literal": "Andrew Keunwoo Jang" }, { "literal": "Jing Yu Koh" }, { "literal": "Ruslan Salakhutdinov" } ], "citation-key": "arxiv_2606_09764", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "495420bd87ba06ae2ec1763865b1145271e964b4bd40145c8e591653a1581c42", "submitted_at": "2026-06-08T17:27:13Z", "updated_at": "2026-06-08T17:27:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.09764v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.09764", "catalog_id": "arxiv:2606.09764", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-08", "identifiers": { "arxiv": "2606.09764" }, "record_updated_at": "2026-06-08" }, "genre": "preprint", "id": "arxiv:2606.09764", "issued": { "date-parts": [ [ 2026, 6, 8 ] ] }, "note": "First published: 2026-06-08; Recorded arXiv version: v1; Recorded version updated: 2026-06-08T17:27:13Z.", "title": "iOSWorld: A Benchmark for Personally Intelligent Phone Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.11042v4", "archive": "arXiv", "archive_location": "2606.11042", "author": [ { "literal": "Liya Zhu" }, { "literal": "Jingzhe Ding" }, { "literal": "Jian Zhang" }, { "literal": "Jianbo Xue" }, { "literal": "Shihao Liang" }, { "literal": "Ge Zhang" }, { "literal": "Yi Zhu" }, { "literal": "Duju Zeng" }, { "literal": "Xiang Gao" }, { "literal": "Qingshui Gu" }, { "literal": "Mailun Gao" }, { "literal": "Huimin Che" }, { "literal": "Yan Zhao" }, { "literal": "Peiheng Zhou" }, { "literal": "Haojun Wang" }, { "literal": "Chaobo Xian" }, { "literal": "Lili Le" }, { "literal": "Chi Wu" }, { "literal": "Yiwei Liu" }, { "literal": "Shengda Long" }, { "literal": "Jiale Yang" }, { "literal": "Fangzhi Xu" }, { "literal": "Sijin Wu" }, { "literal": "Haodong Duan" }, { "literal": "Chao He" }, { "literal": "Zhaojian Li" }, { "literal": "Minchao Wang" }, { "literal": "Huan Zhou" }, { "literal": "Jiani Hou" }, { "literal": "Chuqian Yu" }, { "literal": "Weiran Shi" }, { "literal": "Hongwan Gao" }, { "literal": "Jiamin Chen" }, { "literal": "Guanhong Chen" }, { "literal": "Tingqin Luo" }, { "literal": "Kaiyuan Zhang" }, { "literal": "Zhixin Yao" }, { "literal": "Qing Hua" }, { "literal": "Yuhao Jiang" }, { "literal": "Jin Chen" }, { "literal": "Pu Chen" }, { "literal": "Zhenyu Hu" }, { "literal": "Xingyu Li" }, { "literal": "Zhengxuan Jiang" }, { "literal": "Meng Cao" }, { "literal": "Tianfeng Long" }, { "literal": "Haozhe Wang" }, { "literal": "Mingzhang Wang" }, { "literal": "Yichen Zhang" }, { "literal": "Yiming Dai" }, { "literal": "Chenchen Zhang" }, { "literal": "Jiaying Wang" }, { "literal": "Xinying Liu" }, { "literal": "Xingzu Liu" }, { "literal": "Lingling Zhang" }, { "literal": "Xinjie Chen" }, { "literal": "Yujia Qin" }, { "literal": "Wangchunshu Zhou" }, { "literal": "Zhiyong Wu" }, { "literal": "Yang Liu" }, { "literal": "Jiaheng Liu" }, { "literal": "Lei Zhang" }, { "literal": "Shen Yan" }, { "literal": "Wenhao Huang" }, { "literal": "Zaiyuan Wang" }, { "literal": "Xiaolong Chang" } ], "citation-key": "arxiv_2606_11042", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "037a3197137d27318320a3fd23fb5e606a1aa748133e57213cd02fd10aad4c4d", "submitted_at": "2026-06-09T16:10:16Z", "updated_at": "2026-07-17T12:22:12Z", "version": 4, "versioned_url": "https://arxiv.org/abs/2606.11042v4" }, "canonical_source_url": "https://arxiv.org/abs/2606.11042", "catalog_id": "arxiv:2606.11042", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-09", "identifiers": { "arxiv": "2606.11042" }, "record_updated_at": "2026-07-17" }, "genre": "preprint", "id": "arxiv:2606.11042", "issued": { "date-parts": [ [ 2026, 6, 9 ] ] }, "note": "First published: 2026-06-09; Recorded arXiv version: v4; Recorded version updated: 2026-07-17T12:22:12Z.", "title": "Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields", "type": "article", "version": "4" }, { "URL": "https://arxiv.org/abs/2606.11078v1", "archive": "arXiv", "archive_location": "2606.11078", "author": [ { "literal": "Jaewoo Lee" }, { "literal": "Zaid Khan" }, { "literal": "Archiki Prasad" }, { "literal": "Justin Chih-Yao Chen" }, { "literal": "Supriyo Chakraborty" }, { "literal": "Kartik Balasubramaniam" }, { "literal": "Sambit Sahu" }, { "literal": "Elias Stengel-Eskin" }, { "literal": "Hyunji Lee" }, { "literal": "Mohit Bansal" } ], "citation-key": "arxiv_2606_11078", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a8ed399c8b489a80c5313a00232d2ecca0fbc8228b1f56745c02d4fedb6f6908", "submitted_at": "2026-06-09T16:39:10Z", "updated_at": "2026-06-09T16:39:10Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.11078v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.11078", "catalog_id": "arxiv:2606.11078", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-09", "identifiers": { "arxiv": "2606.11078" }, "record_updated_at": "2026-06-09" }, "genre": "preprint", "id": "arxiv:2606.11078", "issued": { "date-parts": [ [ 2026, 6, 9 ] ] }, "note": "First published: 2026-06-09; Recorded arXiv version: v1; Recorded version updated: 2026-06-09T16:39:10Z.", "title": "A History-Aware Visually Grounded Critic for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.13239v2", "archive": "arXiv", "archive_location": "2606.13239", "author": [ { "literal": "Jiaxin Ai" }, { "literal": "Tao Hu" }, { "literal": "Xuemeng Yang" }, { "literal": "Shu Zou" }, { "literal": "Hairong Zhang" }, { "literal": "Daocheng Fu" }, { "literal": "Yu Yang" }, { "literal": "Hongbin Zhou" }, { "literal": "Nianchen Deng" }, { "literal": "Pinlong Cai" }, { "literal": "Zhongyuan Wang" }, { "literal": "Botian Shi" }, { "literal": "Kaipeng Zhang" }, { "literal": "Licheng Wen" } ], "citation-key": "arxiv_2606_13239", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "edf7e63803efba89551aa297bff1da9b0babf8ec898a2971fa6c8a20cc58dea8", "submitted_at": "2026-06-11T11:53:32Z", "updated_at": "2026-06-30T03:31:20Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.13239v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.13239", "catalog_id": "arxiv:2606.13239", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-06-11", "identifiers": { "arxiv": "2606.13239" }, "record_updated_at": "2026-06-30" }, "genre": "preprint", "id": "arxiv:2606.13239", "issued": { "date-parts": [ [ 2026, 6, 11 ] ] }, "note": "First published: 2026-06-11; Recorded arXiv version: v2; Recorded version updated: 2026-06-30T03:31:20Z.", "title": "ComAct: Reframing Professional Software Manipulation via COM-as-Action Paradigm", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.13385v2", "archive": "arXiv", "archive_location": "2606.13385", "author": [ { "literal": "Zihao Wang" }, { "literal": "Yiming Li" }, { "literal": "Yutong Wu" }, { "literal": "Kangjie Chen" }, { "literal": "Zheyu Liu" }, { "literal": "Fok Kar Wai" }, { "literal": "Pin-Yu Chen" }, { "literal": "Vrizlynn L. L. Thing" }, { "literal": "Bo Li" }, { "literal": "Dacheng Tao" }, { "literal": "Tianwei Zhang" } ], "citation-key": "arxiv_2606_13385", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "88b7e2ee8717661e4f8b417916e35b844171ecd8a8a2add657cf341c8f759536", "submitted_at": "2026-06-11T14:12:43Z", "updated_at": "2026-07-27T14:19:24Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.13385v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.13385", "catalog_id": "arxiv:2606.13385", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-11", "identifiers": { "arxiv": "2606.13385" }, "record_updated_at": "2026-07-27" }, "genre": "preprint", "id": "arxiv:2606.13385", "issued": { "date-parts": [ [ 2026, 6, 11 ] ] }, "note": "First published: 2026-06-11; Recorded arXiv version: v2; Recorded version updated: 2026-07-27T14:19:24Z.", "title": "Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.14106v1", "archive": "arXiv", "archive_location": "2606.14106", "author": [ { "literal": "Seoyoung Choi" }, { "literal": "Minseok Ko" }, { "literal": "Hyunseok Lee" }, { "literal": "Kunwoong Kim" }, { "literal": "Woomin Song" }, { "literal": "Chanseok Jeon" }, { "literal": "Jinwoo Shin" } ], "citation-key": "arxiv_2606_14106", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1fb87da27071ab21519599b7d16359a38118aa358f620de94e66b15cad040828", "submitted_at": "2026-06-12T04:39:01Z", "updated_at": "2026-06-12T04:39:01Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.14106v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.14106", "catalog_id": "arxiv:2606.14106", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-12", "identifiers": { "arxiv": "2606.14106" }, "record_updated_at": "2026-06-12" }, "genre": "preprint", "id": "arxiv:2606.14106", "issued": { "date-parts": [ [ 2026, 6, 12 ] ] }, "note": "First published: 2026-06-12; Recorded arXiv version: v1; Recorded version updated: 2026-06-12T04:39:01Z.", "title": "Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.14832v1", "archive": "arXiv", "archive_location": "2606.14832", "author": [ { "literal": "Chenxin Li" }, { "literal": "Zhengyao Fang" }, { "literal": "Zhengyang Tang" }, { "literal": "Pengyuan Lyu" }, { "literal": "Xingran Zhou" }, { "literal": "Xin Lai" }, { "literal": "Fei Tang" }, { "literal": "Liang Wu" }, { "literal": "Yiduo Guo" }, { "literal": "Weinong Wang" }, { "literal": "Junyi Li" }, { "literal": "Yi Zhang" }, { "literal": "Yang Ding" }, { "literal": "Huawen Shen" }, { "literal": "Sunqi Fan" }, { "literal": "Shangpin Peng" }, { "literal": "Zheng Ruan" }, { "literal": "Anran Zhang" }, { "literal": "Benyou Wang" }, { "literal": "Chengquan Zhang" }, { "literal": "Han Hu" } ], "citation-key": "arxiv_2606_14832", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "280ec12bcb246f1f07cf3627011945e8a9ff4afed93b1fb2937f2d6b1bc1323b", "submitted_at": "2026-06-12T15:01:32Z", "updated_at": "2026-06-12T15:01:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.14832v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.14832", "catalog_id": "arxiv:2606.14832", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-12", "identifiers": { "arxiv": "2606.14832" }, "record_updated_at": "2026-06-12" }, "genre": "preprint", "id": "arxiv:2606.14832", "issued": { "date-parts": [ [ 2026, 6, 12 ] ] }, "note": "First published: 2026-06-12; Recorded arXiv version: v1; Recorded version updated: 2026-06-12T15:01:32Z.", "title": "PhoneHarness: Harnessing Phone-Use Agents through Mixed GUI, CLI, and Tool Actions", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.15017v2", "archive": "arXiv", "archive_location": "2606.15017", "author": [ { "literal": "Sina Hajimiri" }, { "literal": "Masih Aminbeidokhti" }, { "literal": "Jose Dolz" }, { "literal": "Ismail Ben Ayed" }, { "literal": "Issam H. Laradji" }, { "literal": "Spandana Gella" }, { "literal": "Nicolas Gontier" } ], "citation-key": "arxiv_2606_15017", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "50f1acd5727f84946cf95d181f270dbc4f46f15d24339e6cb257fea1b23b21a0", "submitted_at": "2026-06-12T23:30:14Z", "updated_at": "2026-08-30T17:49:13Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.15017v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.15017", "catalog_id": "arxiv:2606.15017", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-12", "identifiers": { "arxiv": "2606.15017" }, "record_updated_at": "2026-08-30" }, "genre": "preprint", "id": "arxiv:2606.15017", "issued": { "date-parts": [ [ 2026, 6, 12 ] ] }, "note": "First published: 2026-06-12; Recorded arXiv version: v2; Recorded version updated: 2026-08-30T17:49:13Z.", "title": "Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.15034v3", "archive": "arXiv", "archive_location": "2606.15034", "author": [ { "literal": "Mina Mohammadmirzaei" }, { "literal": "Jeffrey Flanigan" } ], "citation-key": "arxiv_2606_15034", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7bf0297fba942441b19bcfeba147838bdef888b428f1150c2462d8bf3d986882", "submitted_at": "2026-06-13T00:32:24Z", "updated_at": "2026-10-09T19:45:15Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2606.15034v3" }, "canonical_source_url": "https://arxiv.org/abs/2606.15034", "catalog_id": "arxiv:2606.15034", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-13", "identifiers": { "arxiv": "2606.15034" }, "record_updated_at": "2026-10-09" }, "genre": "preprint", "id": "arxiv:2606.15034", "issued": { "date-parts": [ [ 2026, 6, 13 ] ] }, "note": "First published: 2026-06-13; Recorded arXiv version: v3; Recorded version updated: 2026-10-09T19:45:15Z.", "title": "OSGuard: A Benchmark for Safety in Computer-Use Agents", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2606.16659v1", "archive": "arXiv", "archive_location": "2606.16659", "author": [ { "literal": "Y. H. Zhou" }, { "literal": "Z. M. Ma" }, { "literal": "Y. J. Zhou" }, { "literal": "Y. T. Li" }, { "literal": "H. X. Xiang" }, { "literal": "Y. M. Cheng" }, { "literal": "T. L. Chen" }, { "literal": "K. J. Zhang" }, { "literal": "Z. H. Nan" }, { "literal": "J. H. Ni" }, { "literal": "Z. Wu" }, { "literal": "Q. Y. Pan" }, { "literal": "S. Zhang" }, { "literal": "S. Cheng" }, { "literal": "M. Y. Luo" } ], "citation-key": "arxiv_2606_16659", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6311b22ddb2262c74efbe79acc8856335f6870aa24197c507f3f7b048d6f7901", "submitted_at": "2026-06-15T12:53:55Z", "updated_at": "2026-06-15T12:53:55Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.16659v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.16659", "catalog_id": "arxiv:2606.16659", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-15", "identifiers": { "arxiv": "2606.16659" }, "record_updated_at": "2026-06-15" }, "genre": "preprint", "id": "arxiv:2606.16659", "issued": { "date-parts": [ [ 2026, 6, 15 ] ] }, "note": "First published: 2026-06-15; Recorded arXiv version: v1; Recorded version updated: 2026-06-15T12:53:55Z.", "title": "FraudSMSWalker: Benchmarking Agentic Large Language Models for SMS-to-Webpage Fraud Detection", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.16748v1", "archive": "arXiv", "archive_location": "2606.16748", "author": [ { "literal": "Lawrence Keunho Jang" }, { "literal": "Andrew Keunwoo Jang" }, { "literal": "Jing Yu Koh" }, { "literal": "Ruslan Salakhutdinov" } ], "citation-key": "arxiv_2606_16748", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "acc529f96d6df52e3560f66df8e0630dabdae3ab30cadc8b59123f347eba187a", "submitted_at": "2026-06-15T14:08:09Z", "updated_at": "2026-06-15T14:08:09Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.16748v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.16748", "catalog_id": "arxiv:2606.16748", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-15", "identifiers": { "arxiv": "2606.16748" }, "record_updated_at": "2026-06-15" }, "genre": "preprint", "id": "arxiv:2606.16748", "issued": { "date-parts": [ [ 2026, 6, 15 ] ] }, "note": "First published: 2026-06-15; Recorded arXiv version: v1; Recorded version updated: 2026-06-15T14:08:09Z.", "title": "MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.16802v1", "archive": "arXiv", "archive_location": "2606.16802", "author": [ { "literal": "Anqi Zou" }, { "literal": "Han Deng" }, { "literal": "Chengyu Zhang" }, { "literal": "Junquan Hu" }, { "literal": "Yu Wang" }, { "literal": "Yuxiang Xing" }, { "literal": "Aokai Zhang" }, { "literal": "Hanling Zhang" }, { "literal": "Zhaoyang Liu" }, { "literal": "Ben Fei" }, { "literal": "Zhihui Wang" }, { "literal": "Wanli Ouyang" } ], "citation-key": "arxiv_2606_16802", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7487f31097bc8de90dfbe836051d47f407c26e31a42bd751bedfd38cf4f1742c", "submitted_at": "2026-06-15T14:42:33Z", "updated_at": "2026-06-15T14:42:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.16802v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.16802", "catalog_id": "arxiv:2606.16802", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-15", "identifiers": { "arxiv": "2606.16802" }, "record_updated_at": "2026-06-15" }, "genre": "preprint", "id": "arxiv:2606.16802", "issued": { "date-parts": [ [ 2026, 6, 15 ] ] }, "note": "First published: 2026-06-15; Recorded arXiv version: v1; Recorded version updated: 2026-06-15T14:42:33Z.", "title": "LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.17321v1", "archive": "arXiv", "archive_location": "2606.17321", "author": [ { "literal": "Jaehun Jung" }, { "literal": "Ximing Lu" }, { "literal": "Brandon Cui" }, { "literal": "Muhammad Khalifa" }, { "literal": "Shaokun Zhang" }, { "literal": "Hao Zhang" }, { "literal": "Jin Xu" }, { "literal": "Amala Sanjay Deshmukh" }, { "literal": "Karan Sapra" }, { "literal": "Andrew Tao" }, { "literal": "Yejin Choi" }, { "literal": "Jan Kautz" }, { "literal": "Mingjie Liu" }, { "literal": "Yi Dong" } ], "citation-key": "arxiv_2606_17321", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d7e6e0b66f2923cb66ab7aa404bedf9f4e1c73ae00df2e9fbfa7a5c3795cac4c", "submitted_at": "2026-06-15T22:04:11Z", "updated_at": "2026-06-15T22:04:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.17321v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.17321", "catalog_id": "arxiv:2606.17321", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-15", "identifiers": { "arxiv": "2606.17321" }, "record_updated_at": "2026-06-15" }, "genre": "preprint", "id": "arxiv:2606.17321", "issued": { "date-parts": [ [ 2026, 6, 15 ] ] }, "note": "First published: 2026-06-15; Recorded arXiv version: v1; Recorded version updated: 2026-06-15T22:04:11Z.", "title": "ProCUA-SFT Technical Report", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.18101v2", "archive": "arXiv", "archive_location": "2606.18101", "author": [ { "literal": "Jingyuan Huang" }, { "literal": "Zuming Huang" }, { "literal": "Yucheng Shi" }, { "literal": "Tianze Yang" }, { "literal": "Xiaoming Zhai" }, { "literal": "Wei Chu" }, { "literal": "Ninghao Liu" } ], "citation-key": "arxiv_2606_18101", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "49b935d8993be0d03ae778375f908b68b3ec3c614a7406b6b3f6a00f2a89ef1f", "submitted_at": "2026-06-16T16:02:41Z", "updated_at": "2026-06-17T08:58:52Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.18101v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.18101", "catalog_id": "arxiv:2606.18101", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-16", "identifiers": { "arxiv": "2606.18101" }, "record_updated_at": "2026-06-17" }, "genre": "preprint", "id": "arxiv:2606.18101", "issued": { "date-parts": [ [ 2026, 6, 16 ] ] }, "note": "First published: 2026-06-16; Recorded arXiv version: v2; Recorded version updated: 2026-06-17T08:58:52Z.", "title": "Trust the Right Teacher: Quality-Aware Self-Distillation for GUI Grounding", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.18671v1", "archive": "arXiv", "archive_location": "2606.18671", "author": [ { "literal": "Yujin Zhang" }, { "literal": "Daye Nam" } ], "citation-key": "arxiv_2606_18671", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c52695e860206d02735839853416965ec3d42fac134ebfa0e8f2f5f5ea043914", "submitted_at": "2026-06-17T04:13:20Z", "updated_at": "2026-06-17T04:13:20Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.18671v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.18671", "catalog_id": "arxiv:2606.18671", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-17", "identifiers": { "arxiv": "2606.18671" }, "record_updated_at": "2026-06-17" }, "genre": "preprint", "id": "arxiv:2606.18671", "issued": { "date-parts": [ [ 2026, 6, 17 ] ] }, "note": "First published: 2026-06-17; Recorded arXiv version: v1; Recorded version updated: 2026-06-17T04:13:20Z.", "title": "HANSEL: Extracting Breadcrumbs from Web Agent Trajectories for Interactive Verification", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.19388v1", "archive": "arXiv", "archive_location": "2606.19388", "author": [ { "literal": "Li Gu" }, { "literal": "Zihuan Jiang" }, { "literal": "Linqiang Guo" }, { "literal": "Zhixiang Chi" }, { "literal": "Ziqiang Wang" }, { "literal": "Huan Liu" }, { "literal": "Yuanhao Yu" }, { "literal": "Tse-Hsun Chen" }, { "literal": "Yang Wang" } ], "citation-key": "arxiv_2606_19388", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9bf8b07e08b275d90fcbfc63d867f54f5579509132cc9cf694d01447b7f91b78", "submitted_at": "2026-06-16T02:36:22Z", "updated_at": "2026-06-16T02:36:22Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.19388v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.19388", "catalog_id": "arxiv:2606.19388", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-06-16", "identifiers": { "arxiv": "2606.19388" }, "record_updated_at": "2026-06-16" }, "genre": "preprint", "id": "arxiv:2606.19388", "issued": { "date-parts": [ [ 2026, 6, 16 ] ] }, "note": "First published: 2026-06-16; Recorded arXiv version: v1; Recorded version updated: 2026-06-16T02:36:22Z.", "title": "Beyond the GUI Paradigm: Do Mobile Agents Need the Phone Screen?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.19930v2", "archive": "arXiv", "archive_location": "2606.19930", "author": [ { "literal": "Guangyi Liu" }, { "literal": "Pengxiang Zhao" }, { "literal": "Gao Wu" }, { "literal": "Yiwen Yin" }, { "literal": "Mading Li" }, { "literal": "Liang Liu" }, { "literal": "Congxiao Liu" }, { "literal": "Zhang Qi" }, { "literal": "Mengyan Wang" }, { "literal": "Liang Guo" }, { "literal": "Jiangning Zhang" }, { "literal": "Yong Liu" } ], "citation-key": "arxiv_2606_19930", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cab809618ea4e218da26b4ae11bf937859c3a4bc240c91dd27b3a0d333737069", "submitted_at": "2026-06-18T08:29:33Z", "updated_at": "2026-08-07T03:54:59Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.19930v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.19930", "catalog_id": "arxiv:2606.19930", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-18", "identifiers": { "arxiv": "2606.19930" }, "record_updated_at": "2026-08-07" }, "genre": "preprint", "id": "arxiv:2606.19930", "issued": { "date-parts": [ [ 2026, 6, 18 ] ] }, "note": "First published: 2026-06-18; Recorded arXiv version: v2; Recorded version updated: 2026-08-07T03:54:59Z.", "title": "MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.20363v1", "archive": "arXiv", "archive_location": "2606.20363", "author": [ { "literal": "Yuexing Hao" }, { "literal": "Xiaomin Li" } ], "citation-key": "arxiv_2606_20363", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b370491b37d5bf25c624f2bac654e438459c8084b5da292fb18a0bf9142adaa2", "submitted_at": "2026-06-18T15:25:42Z", "updated_at": "2026-06-18T15:25:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.20363v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.20363", "catalog_id": "arxiv:2606.20363", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-18", "identifiers": { "arxiv": "2606.20363" }, "record_updated_at": "2026-06-18" }, "genre": "preprint", "id": "arxiv:2606.20363", "issued": { "date-parts": [ [ 2026, 6, 18 ] ] }, "note": "First published: 2026-06-18; Recorded arXiv version: v1; Recorded version updated: 2026-06-18T15:25:42Z.", "title": "Automating SKILL.md Generation for Computer-Using Agents via Interaction Trajectory Mining", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.20622v1", "archive": "arXiv", "archive_location": "2606.20622", "author": [ { "literal": "Daniel Beechey" }, { "literal": "Derek Yuen" }, { "literal": "Jianheng Liu" }, { "literal": "Dezhao Luo" }, { "literal": "Tiantian He" }, { "literal": "Weilin Luo" }, { "literal": "Jun Wang" }, { "literal": "Kun Shao" } ], "citation-key": "arxiv_2606_20622", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9e761b20caa6882605ecfd6cd001aeda4a2d36b2e702eea9a6d70451fc0ea9eb", "submitted_at": "2026-05-26T13:32:14Z", "updated_at": "2026-05-26T13:32:14Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.20622v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.20622", "catalog_id": "arxiv:2606.20622", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-05-26", "identifiers": { "arxiv": "2606.20622" }, "record_updated_at": "2026-05-26" }, "genre": "preprint", "id": "arxiv:2606.20622", "issued": { "date-parts": [ [ 2026, 5, 26 ] ] }, "note": "First published: 2026-05-26; Recorded arXiv version: v1; Recorded version updated: 2026-05-26T13:32:14Z.", "title": "Darwin Mobile Agent: A Roadmap for Self-Evolution", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.20717v2", "archive": "arXiv", "archive_location": "2606.20717", "author": [ { "literal": "Xuelong Dai" }, { "literal": "Jianyu Ma" }, { "literal": "Boyang Ma" }, { "literal": "Biwei Yan" }, { "literal": "Yijun Yang" }, { "literal": "Yue Zhang" } ], "citation-key": "arxiv_2606_20717", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fea441692131ec677e4a9ae54f9ecef1f31d422e20884b496b124e99392e59c1", "submitted_at": "2026-06-16T15:31:33Z", "updated_at": "2026-10-15T04:29:19Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.20717v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.20717", "catalog_id": "arxiv:2606.20717", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-16", "identifiers": { "arxiv": "2606.20717" }, "record_updated_at": "2026-10-15" }, "genre": "preprint", "id": "arxiv:2606.20717", "issued": { "date-parts": [ [ 2026, 6, 16 ] ] }, "note": "First published: 2026-06-16; Recorded arXiv version: v2; Recorded version updated: 2026-10-15T04:29:19Z.", "title": "MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.20785v2", "archive": "arXiv", "archive_location": "2606.20785", "author": [ { "literal": "Ahmed Awadallah" }, { "literal": "Sahil Gupta" }, { "literal": "Yash Lara" }, { "literal": "Yadong Lu" }, { "literal": "Hussein Mozannar" }, { "literal": "Akshay Nambi" }, { "literal": "Zach Nussbaum" }, { "literal": "Yash Pandya" }, { "literal": "Aravind Rajeswaran" }, { "literal": "Corby Rosset" }, { "literal": "Alexey Taymanov" }, { "literal": "Luiz do Valle" }, { "literal": "Vibhav Vineet" }, { "literal": "Spencer Whitehead" }, { "literal": "Andrew Zhao" } ], "citation-key": "arxiv_2606_20785", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "414eced85da9f76828e4398b977f9b8dac77236c97d1594108c8f5dfc3be7cbb", "submitted_at": "2026-06-18T17:53:03Z", "updated_at": "2026-07-22T17:47:01Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.20785v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.20785", "catalog_id": "arxiv:2606.20785", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-18", "identifiers": { "arxiv": "2606.20785" }, "record_updated_at": "2026-07-22" }, "genre": "preprint", "id": "arxiv:2606.20785", "issued": { "date-parts": [ [ 2026, 6, 18 ] ] }, "note": "First published: 2026-06-18; Recorded arXiv version: v2; Recorded version updated: 2026-07-22T17:47:01Z.", "title": "Fara-1.5: Scalable Learning Environments for Computer Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.20910v1", "archive": "arXiv", "archive_location": "2606.20910", "author": [ { "literal": "Dayeon Kang" }, { "literal": "Hyejun Jeong" }, { "literal": "Jade Sheffey" }, { "literal": "Pubali Datta" }, { "literal": "Amir Houmansadr" } ], "citation-key": "arxiv_2606_20910", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8b2c6080f0de11c06fe73c30195c7c8529e07de283dc45fa47709072846d385a", "submitted_at": "2026-06-18T20:07:39Z", "updated_at": "2026-06-18T20:07:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.20910v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.20910", "catalog_id": "arxiv:2606.20910", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-18", "identifiers": { "arxiv": "2606.20910" }, "record_updated_at": "2026-06-18" }, "genre": "preprint", "id": "arxiv:2606.20910", "issued": { "date-parts": [ [ 2026, 6, 18 ] ] }, "note": "First published: 2026-06-18; Recorded arXiv version: v1; Recorded version updated: 2026-06-18T20:07:39Z.", "title": "Whose Agent Are You? Multi-Layer Fingerprinting and Attribution of Autonomous Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.21654v2", "archive": "arXiv", "archive_location": "2606.21654", "author": [ { "literal": "Vincent Siu" }, { "literal": "Manasi Sharma" }, { "literal": "Dawn Song" }, { "literal": "Daniel Yue Zhang" }, { "literal": "Chenguang Wang" }, { "literal": "Ying Liu" } ], "citation-key": "arxiv_2606_21654", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "69e668a9491a31e8d8849bab8cc63eede9878fff10ccb0ae0a211610e90819b3", "submitted_at": "2026-06-19T18:00:14Z", "updated_at": "2026-08-19T17:29:04Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.21654v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.21654", "catalog_id": "arxiv:2606.21654", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-19", "identifiers": { "arxiv": "2606.21654" }, "record_updated_at": "2026-08-19" }, "genre": "preprint", "id": "arxiv:2606.21654", "issued": { "date-parts": [ [ 2026, 6, 19 ] ] }, "note": "First published: 2026-06-19; Recorded arXiv version: v2; Recorded version updated: 2026-08-19T17:29:04Z.", "title": "ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.22557v1", "archive": "arXiv", "archive_location": "2606.22557", "author": [ { "literal": "Yikun Fu" }, { "literal": "Bowen Fu" }, { "literal": "Zhenyu Wu" }, { "literal": "Shuang Cheng" }, { "literal": "Xiaowei Sun" }, { "literal": "Bowen Yang" }, { "literal": "Zehao Li" }, { "literal": "Yibo Zhao" }, { "literal": "Zichen Ding" }, { "literal": "Zhoumianze Liu" }, { "literal": "Shijie Wang" }, { "literal": "Biqing Qi" }, { "literal": "Bowen Zhou" } ], "citation-key": "arxiv_2606_22557", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6dc52f58bb2237d12041fa8b994b20bd1e41290aaab0c70d2c5dd79fcaca37b6", "submitted_at": "2026-06-21T15:32:57Z", "updated_at": "2026-06-21T15:32:57Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.22557v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.22557", "catalog_id": "arxiv:2606.22557", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-21", "identifiers": { "arxiv": "2606.22557" }, "record_updated_at": "2026-06-21" }, "genre": "preprint", "id": "arxiv:2606.22557", "issued": { "date-parts": [ [ 2026, 6, 21 ] ] }, "note": "First published: 2026-06-21; Recorded arXiv version: v1; Recorded version updated: 2026-06-21T15:32:57Z.", "title": "MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.22948v1", "archive": "arXiv", "archive_location": "2606.22948", "author": [ { "literal": "Yincheng Zhou" }, { "literal": "Athena Zhuoming Zhong" }, { "literal": "Shijie Zhang" }, { "literal": "Kevin Zhang" }, { "literal": "Teresa Xiaotao Shang" }, { "literal": "Shanghang Zhang" } ], "citation-key": "arxiv_2606_22948", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fbd8892e06a5e847c16aa3101b98b3d669a783b84ee8e3d4d4d810754914028b", "submitted_at": "2026-06-22T07:28:55Z", "updated_at": "2026-06-22T07:28:55Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.22948v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.22948", "catalog_id": "arxiv:2606.22948", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-22", "identifiers": { "arxiv": "2606.22948" }, "record_updated_at": "2026-06-22" }, "genre": "preprint", "id": "arxiv:2606.22948", "issued": { "date-parts": [ [ 2026, 6, 22 ] ] }, "note": "First published: 2026-06-22; Recorded arXiv version: v1; Recorded version updated: 2026-06-22T07:28:55Z.", "title": "ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.23189v2", "archive": "arXiv", "archive_location": "2606.23189", "author": [ { "literal": "Anmol Goel" }, { "literal": "Iryna Gurevych" } ], "citation-key": "arxiv_2606_23189", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "da8190bc5afc0bae400fc8ab5e7d80a7f34488c3400dce4477ca61635b19d5b7", "submitted_at": "2026-06-22T11:36:58Z", "updated_at": "2026-09-10T18:49:54Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.23189v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.23189", "catalog_id": "arxiv:2606.23189", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-22", "identifiers": { "arxiv": "2606.23189" }, "record_updated_at": "2026-09-10" }, "genre": "preprint", "id": "arxiv:2606.23189", "issued": { "date-parts": [ [ 2026, 6, 22 ] ] }, "note": "First published: 2026-06-22; Recorded arXiv version: v2; Recorded version updated: 2026-09-10T18:49:54Z.", "title": "Capable but Careless: Do Computer-Use Agents Follow Contextual Integrity?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.24515v1", "archive": "arXiv", "archive_location": "2606.24515", "author": [ { "literal": "Marta Sumyk" }, { "literal": "Oleksandr Kosovan" } ], "citation-key": "arxiv_2606_24515", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0aeb8c28b570133f575c2f336b6dab0dcbecdfaec78549115407b21722816e8a", "submitted_at": "2026-06-23T12:46:29Z", "updated_at": "2026-06-23T12:46:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.24515v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.24515", "catalog_id": "arxiv:2606.24515", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-23", "identifiers": { "arxiv": "2606.24515" }, "record_updated_at": "2026-06-23" }, "genre": "preprint", "id": "arxiv:2606.24515", "issued": { "date-parts": [ [ 2026, 6, 23 ] ] }, "note": "First published: 2026-06-23; Recorded arXiv version: v1; Recorded version updated: 2026-06-23T12:46:29Z.", "title": "Reinforcement Learning for Computer-Use Agents with Autonomous Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.24525v1", "archive": "arXiv", "archive_location": "2606.24525", "author": [ { "literal": "Jiachen Qian" } ], "citation-key": "arxiv_2606_24525", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a3d0341f41a8ecba57fb7ed9e4b266bb4a176639c742be2b4fc98555754f5744", "submitted_at": "2026-06-23T12:57:02Z", "updated_at": "2026-06-23T12:57:02Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.24525v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.24525", "catalog_id": "arxiv:2606.24525", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-23", "identifiers": { "arxiv": "2606.24525" }, "record_updated_at": "2026-06-23" }, "genre": "preprint", "id": "arxiv:2606.24525", "issued": { "date-parts": [ [ 2026, 6, 23 ] ] }, "note": "First published: 2026-06-23; Recorded arXiv version: v1; Recorded version updated: 2026-06-23T12:57:02Z.", "title": "VisCritic: Visual State Comparison as Process Reward for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.24551v1", "archive": "arXiv", "archive_location": "2606.24551", "author": [ { "literal": "Xiao Zhou" }, { "literal": "Siyue Zhang" }, { "literal": "Yilun Zhao" }, { "literal": "Jinbiao Wei" }, { "literal": "Tingyu Song" }, { "literal": "Arman Cohan" }, { "literal": "Chen Zhao" } ], "citation-key": "arxiv_2606_24551", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "77f777f05b1c169ee4ad7ae7d77e9d1fc80a0696e9a1f6eefb5028a5c3e6fe5d", "submitted_at": "2026-06-22T17:05:07Z", "updated_at": "2026-06-22T17:05:07Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.24551v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.24551", "catalog_id": "arxiv:2606.24551", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-22", "identifiers": { "arxiv": "2606.24551" }, "record_updated_at": "2026-06-22" }, "genre": "preprint", "id": "arxiv:2606.24551", "issued": { "date-parts": [ [ 2026, 6, 22 ] ] }, "note": "First published: 2026-06-22; Recorded arXiv version: v1; Recorded version updated: 2026-06-22T17:05:07Z.", "title": "GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.25760v2", "archive": "arXiv", "archive_location": "2606.25760", "author": [ { "literal": "Divake Kumar" }, { "literal": "Sina Tayebati" }, { "literal": "Devashri Naik" }, { "literal": "Amanda Sofie Rios" }, { "literal": "Nilesh Ahuja" }, { "literal": "Omesh Tickoo" }, { "literal": "Ranganath Krishnan" }, { "literal": "Amit Ranjan Trivedi" } ], "citation-key": "arxiv_2606_25760", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "628d08bdbf23ae9e599bf767f0503cecedcde72dfbfcc183f9ea9f044c6f2d44", "submitted_at": "2026-06-24T12:34:28Z", "updated_at": "2026-09-29T20:55:51Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.25760v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.25760", "catalog_id": "arxiv:2606.25760", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-24", "identifiers": { "arxiv": "2606.25760" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2606.25760", "issued": { "date-parts": [ [ 2026, 6, 24 ] ] }, "note": "First published: 2026-06-24; Recorded arXiv version: v2; Recorded version updated: 2026-09-29T20:55:51Z.", "title": "Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.28480v1", "archive": "arXiv", "archive_location": "2606.28480", "author": [ { "literal": "Shoufa Chen" }, { "literal": "Luyuan Wang" }, { "literal": "Xuan Yang" }, { "literal": "Zhiheng Liu" }, { "literal": "Yuren Cong" }, { "literal": "Yuanfeng Ji" }, { "literal": "Feiyan Zhou" }, { "literal": "Xiaohui Zhang" }, { "literal": "Fanny Yang" }, { "literal": "Belinda Zeng" } ], "citation-key": "arxiv_2606_28480", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6164c42541650e5120adc881b19524cb3b268f08d6eb77fbe91d8c1a02823786", "submitted_at": "2026-06-26T17:59:51Z", "updated_at": "2026-06-26T17:59:51Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.28480v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.28480", "catalog_id": "arxiv:2606.28480", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-06-26", "identifiers": { "arxiv": "2606.28480" }, "record_updated_at": "2026-06-26" }, "genre": "preprint", "id": "arxiv:2606.28480", "issued": { "date-parts": [ [ 2026, 6, 26 ] ] }, "note": "First published: 2026-06-26; Recorded arXiv version: v1; Recorded version updated: 2026-06-26T17:59:51Z.", "title": "TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.29472v1", "archive": "arXiv", "archive_location": "2606.29472", "author": [ { "literal": "Bojie Li" }, { "literal": "Noah Shi" } ], "citation-key": "arxiv_2606_29472", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "54de4e93f4f3e39d904b4661a193765e02c29f8ec7787264aee5616579c32210", "submitted_at": "2026-06-28T15:59:31Z", "updated_at": "2026-06-28T15:59:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.29472v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.29472", "catalog_id": "arxiv:2606.29472", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-28", "identifiers": { "arxiv": "2606.29472" }, "record_updated_at": "2026-06-28" }, "genre": "preprint", "id": "arxiv:2606.29472", "issued": { "date-parts": [ [ 2026, 6, 28 ] ] }, "note": "First published: 2026-06-28; Recorded arXiv version: v1; Recorded version updated: 2026-06-28T15:59:31Z.", "title": "Agent-Computer Observation Interfaces Enable Dynamic Computer Use", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.29537v2", "archive": "arXiv", "archive_location": "2606.29537", "author": [ { "literal": "Mengqi Yuan" }, { "literal": "Zilong Zhou" }, { "literal": "Xinzhuang Xiong" }, { "literal": "Weiming Wu" }, { "literal": "Jiayang Sun" }, { "literal": "Jiamin Song" }, { "literal": "Kaiqian Cui" }, { "literal": "Bowen Wang" }, { "literal": "Haoyuan Wu" }, { "literal": "Yitong Li" }, { "literal": "Dunjie Lu" }, { "literal": "Haikong Lu" }, { "literal": "Qi Zhen" }, { "literal": "Xinyuan Wang" }, { "literal": "Jiaqi Deng" }, { "literal": "Yuhao Yang" }, { "literal": "Cheng Chen" }, { "literal": "Boyuan Zheng" }, { "literal": "Alex Su" }, { "literal": "Xiao Yu" }, { "literal": "Hao Zou" }, { "literal": "Saaket Agashe" }, { "literal": "Xing Han Lu" }, { "literal": "Manpreet Kaur" }, { "literal": "Zhengyang Qi" }, { "literal": "Vincent Sunn Chen" }, { "literal": "Frederic Sala" }, { "literal": "Dayiheng Liu" }, { "literal": "Junyang Lin" }, { "literal": "Zhou Yu" }, { "literal": "Yu Su" }, { "literal": "Siva Reddy" }, { "literal": "Xin Eric Wang" }, { "literal": "Peng Qi" }, { "literal": "Tianbao Xie" }, { "literal": "Tao Yu" } ], "citation-key": "arxiv_2606_29537", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "03035fcbc6cd3e6bf40f244870fb35694ba985ff7f809da66e2a376cbf99d66b", "submitted_at": "2026-06-28T17:59:17Z", "updated_at": "2026-07-13T10:30:17Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.29537v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.29537", "catalog_id": "arxiv:2606.29537", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-28", "identifiers": { "arxiv": "2606.29537" }, "record_updated_at": "2026-07-13" }, "genre": "preprint", "id": "arxiv:2606.29537", "issued": { "date-parts": [ [ 2026, 6, 28 ] ] }, "note": "First published: 2026-06-28; Recorded arXiv version: v2; Recorded version updated: 2026-07-13T10:30:17Z.", "title": "OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2606.30084v1", "archive": "arXiv", "archive_location": "2606.30084", "author": [ { "literal": "Chen Liu" }, { "literal": "Ling Chen" }, { "literal": "Hanzhang Zhou" }, { "literal": "Liangyu Chen" }, { "literal": "Chenglin Cai" }, { "literal": "Xin Yu" }, { "literal": "Steven Hoi" }, { "literal": "Yue Wang" } ], "citation-key": "arxiv_2606_30084", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "50e44e08750d4152f9b5408885f8a6a016c5f789af9712e447ee147fb25adb8d", "submitted_at": "2026-06-29T10:20:39Z", "updated_at": "2026-06-29T10:20:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.30084v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.30084", "catalog_id": "arxiv:2606.30084", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-29", "identifiers": { "arxiv": "2606.30084" }, "record_updated_at": "2026-06-29" }, "genre": "preprint", "id": "arxiv:2606.30084", "issued": { "date-parts": [ [ 2026, 6, 29 ] ] }, "note": "First published: 2026-06-29; Recorded arXiv version: v1; Recorded version updated: 2026-06-29T10:20:39Z.", "title": "One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.30119v1", "archive": "arXiv", "archive_location": "2606.30119", "author": [ { "literal": "Iliana Fayolle" }, { "literal": "Sihem Bouhenniche" }, { "literal": "Samuel Pélissier" }, { "literal": "Pierre Laperdrix" }, { "literal": "Clémentine Maurice" }, { "literal": "Walter Rudametkin" } ], "citation-key": "arxiv_2606_30119", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "770c901e3490b2a43b02296136825e8e6bd035fb7494ea74b3a9af74be1afed1", "submitted_at": "2026-06-29T10:51:13Z", "updated_at": "2026-06-29T10:51:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.30119v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.30119", "catalog_id": "arxiv:2606.30119", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-29", "identifiers": { "arxiv": "2606.30119" }, "record_updated_at": "2026-06-29" }, "genre": "preprint", "id": "arxiv:2606.30119", "issued": { "date-parts": [ [ 2026, 6, 29 ] ] }, "note": "First published: 2026-06-29; Recorded arXiv version: v1; Recorded version updated: 2026-06-29T10:51:13Z.", "title": "On the Internet, Nobody Knows You're an LLM Bot: Unmasking Web Agents with Multi-Layer Fingerprinting", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.31270v1", "archive": "arXiv", "archive_location": "2606.31270", "author": [ { "literal": "Xueqiao Sun" }, { "literal": "Xiaohan Wang" }, { "literal": "Ludwig Schmidt" }, { "literal": "Serena Yeung-Levy" }, { "literal": "Yuhui Zhang" } ], "citation-key": "arxiv_2606_31270", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8911abdc0c9c481d82abb213d368c16fba9fee909c354e0bc6272e14c59abfb7", "submitted_at": "2026-06-30T07:44:37Z", "updated_at": "2026-06-30T07:44:37Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2606.31270v1" }, "canonical_source_url": "https://arxiv.org/abs/2606.31270", "catalog_id": "arxiv:2606.31270", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-06-30", "identifiers": { "arxiv": "2606.31270" }, "record_updated_at": "2026-06-30" }, "genre": "preprint", "id": "arxiv:2606.31270", "issued": { "date-parts": [ [ 2026, 6, 30 ] ] }, "note": "First published: 2026-06-30; Recorded arXiv version: v1; Recorded version updated: 2026-06-30T07:44:37Z.", "title": "Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2606.31980v2", "archive": "arXiv", "archive_location": "2606.31980", "author": [ { "literal": "Meng Chen" }, { "literal": "Anya Ji" }, { "literal": "Tsung-Han Wu" }, { "literal": "Tobias Maringgele" }, { "literal": "David M. Chan" }, { "literal": "Alane Suhr" }, { "literal": "Amy Pavel" } ], "citation-key": "arxiv_2606_31980", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a0a3e641e0a71051f5fbd20f57af3db06f2b65c61025e78adcdd122f773d034b", "submitted_at": "2026-06-30T17:18:47Z", "updated_at": "2026-09-01T04:17:49Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2606.31980v2" }, "canonical_source_url": "https://arxiv.org/abs/2606.31980", "catalog_id": "arxiv:2606.31980", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-06-30", "identifiers": { "arxiv": "2606.31980" }, "record_updated_at": "2026-09-01" }, "genre": "preprint", "id": "arxiv:2606.31980", "issued": { "date-parts": [ [ 2026, 6, 30 ] ] }, "note": "First published: 2026-06-30; Recorded arXiv version: v2; Recorded version updated: 2026-09-01T04:17:49Z.", "title": "DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2607.04334v2", "archive": "arXiv", "archive_location": "2607.04334", "author": [ { "literal": "Guijia Zhang" }, { "literal": "Yuxun Chen" }, { "literal": "Yuheng Qi" }, { "literal": "Harry Yang" } ], "citation-key": "arxiv_2607_04334", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e7fdae5d6e5ea17639fe0e2b8e2d655df7201a438d84acdeb3e0d02a78051ae2", "submitted_at": "2026-07-05T14:33:29Z", "updated_at": "2026-07-18T16:56:38Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2607.04334v2" }, "canonical_source_url": "https://arxiv.org/abs/2607.04334", "catalog_id": "arxiv:2607.04334", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-05", "identifiers": { "arxiv": "2607.04334" }, "record_updated_at": "2026-07-18" }, "genre": "preprint", "id": "arxiv:2607.04334", "issued": { "date-parts": [ [ 2026, 7, 5 ] ] }, "note": "First published: 2026-07-05; Recorded arXiv version: v2; Recorded version updated: 2026-07-18T16:56:38Z.", "title": "Do GUI Agents Believe Their Eyes? Diagnosing State-Belief Reliance on Pixels versus Structure", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2607.05120v1", "archive": "arXiv", "archive_location": "2607.05120", "author": [ { "literal": "Woohyuk Choi" }, { "literal": "Juhee Kim" }, { "literal": "Taehyun Kang" }, { "literal": "Jihyeon Jeong" }, { "literal": "Luyi Xing" }, { "literal": "Byoungyoung Lee" } ], "citation-key": "arxiv_2607_05120", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c5fb30b8be2c6e452471e6398720291ec59a2c5ef96e64baca0735fb01b7a12e", "submitted_at": "2026-07-06T14:07:49Z", "updated_at": "2026-07-06T14:07:49Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.05120v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.05120", "catalog_id": "arxiv:2607.05120", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-06", "identifiers": { "arxiv": "2607.05120" }, "record_updated_at": "2026-07-06" }, "genre": "preprint", "id": "arxiv:2607.05120", "issued": { "date-parts": [ [ 2026, 7, 6 ] ] }, "note": "First published: 2026-07-06; Recorded arXiv version: v1; Recorded version updated: 2026-07-06T14:07:49Z.", "title": "Agent Data Injection Attacks are Realistic Threats to AI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.06118v1", "archive": "arXiv", "archive_location": "2607.06118", "author": [ { "literal": "Wei Dong" }, { "literal": "Tianyu Fu" }, { "literal": "Zhe Yu" }, { "literal": "Hanning Wang" }, { "literal": "Anyang Su" }, { "literal": "Zhizhou Fang" }, { "literal": "Yuyang Chen" }, { "literal": "Shuo Wang" }, { "literal": "Minghui Wu" }, { "literal": "Ping Jiang" }, { "literal": "Zhen Lei" }, { "literal": "Chenxu Zhao" } ], "citation-key": "arxiv_2607_06118", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cc6d846f43b9992388e94f3bcfdfc9a25d09b317fea62336aa22951c8fce9f8b", "submitted_at": "2026-07-07T10:27:31Z", "updated_at": "2026-07-07T10:27:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.06118v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.06118", "catalog_id": "arxiv:2607.06118", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-07", "identifiers": { "arxiv": "2607.06118" }, "record_updated_at": "2026-07-07" }, "genre": "preprint", "id": "arxiv:2607.06118", "issued": { "date-parts": [ [ 2026, 7, 7 ] ] }, "note": "First published: 2026-07-07; Recorded arXiv version: v1; Recorded version updated: 2026-07-07T10:27:31Z.", "title": "WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.10079v3", "archive": "arXiv", "archive_location": "2607.10079", "author": [ { "literal": "Chengguang Gan" }, { "literal": "Hanjun Wei" }, { "literal": "Yunhao Liang" }, { "literal": "Zhixi Cai" }, { "literal": "Qinghao Zhang" }, { "literal": "Shiwen Ni" } ], "citation-key": "arxiv_2607_10079", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "37cfa436d182ed8fbc8823213b9173fbfcf9fb421e8d2ae89a093a5393c374ba", "submitted_at": "2026-07-11T02:08:46Z", "updated_at": "2026-07-16T03:28:43Z", "version": 3, "versioned_url": "https://arxiv.org/abs/2607.10079v3" }, "canonical_source_url": "https://arxiv.org/abs/2607.10079", "catalog_id": "arxiv:2607.10079", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-11", "identifiers": { "arxiv": "2607.10079" }, "record_updated_at": "2026-07-16" }, "genre": "preprint", "id": "arxiv:2607.10079", "issued": { "date-parts": [ [ 2026, 7, 11 ] ] }, "note": "First published: 2026-07-11; Recorded arXiv version: v3; Recorded version updated: 2026-07-16T03:28:43Z.", "title": "MAG: A Web-Agent Benchmark and Harness for Multimodal Action and Guide Generation", "type": "article", "version": "3" }, { "URL": "https://arxiv.org/abs/2607.10608v1", "archive": "arXiv", "archive_location": "2607.10608", "author": [ { "literal": "Yixiong Chen" }, { "literal": "Xinyi Bai" }, { "literal": "Alan Yuille" } ], "citation-key": "arxiv_2607_10608", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0473ced7881ed2007aa49c2767860c33719d5bf1c798dfe4771a4061b13d655e", "submitted_at": "2026-07-12T07:05:31Z", "updated_at": "2026-07-12T07:05:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.10608v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.10608", "catalog_id": "arxiv:2607.10608", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-12", "identifiers": { "arxiv": "2607.10608" }, "record_updated_at": "2026-07-12" }, "genre": "preprint", "id": "arxiv:2607.10608", "issued": { "date-parts": [ [ 2026, 7, 12 ] ] }, "note": "First published: 2026-07-12; Recorded arXiv version: v1; Recorded version updated: 2026-07-12T07:05:31Z.", "title": "The Compliance Trap: Diagnosing How AI Agents Consume Conflicting Memory", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.12640v1", "archive": "arXiv", "archive_location": "2607.12640", "author": [ { "literal": "Chengguang Gan" }, { "literal": "Zhixi Cai" }, { "literal": "Yunhao Liang" }, { "literal": "Hanjun Wei" }, { "literal": "Shiwen Ni" }, { "literal": "Qinghao Zhang" } ], "citation-key": "arxiv_2607_12640", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e96ea0cf71ff705c064db1bcd1d2cbf8b8cd712bb4e8030931cafbb31ff29b10", "submitted_at": "2026-07-14T11:17:35Z", "updated_at": "2026-07-14T11:17:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.12640v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.12640", "catalog_id": "arxiv:2607.12640", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-14", "identifiers": { "arxiv": "2607.12640" }, "record_updated_at": "2026-07-14" }, "genre": "preprint", "id": "arxiv:2607.12640", "issued": { "date-parts": [ [ 2026, 7, 14 ] ] }, "note": "First published: 2026-07-14; Recorded arXiv version: v1; Recorded version updated: 2026-07-14T11:17:35Z.", "title": "A Learning-Rate-Gated Failure of GRPO in a Small Language and Vision-Language Model Web Agent: A Controlled Null and Its Mechanism", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.14443v1", "archive": "arXiv", "archive_location": "2607.14443", "author": [ { "literal": "Yong Liu" }, { "literal": "Zhenyi Zhong" }, { "literal": "Zhanpeng Shi" } ], "citation-key": "arxiv_2607_14443", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a779b2c1ff9ba3b531d399204b0986e3affcea81c67fdd7951e3b7da343b0e17", "submitted_at": "2026-07-16T00:35:25Z", "updated_at": "2026-07-16T00:35:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.14443v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.14443", "catalog_id": "arxiv:2607.14443", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-16", "identifiers": { "arxiv": "2607.14443" }, "record_updated_at": "2026-07-16" }, "genre": "preprint", "id": "arxiv:2607.14443", "issued": { "date-parts": [ [ 2026, 7, 16 ] ] }, "note": "First published: 2026-07-16; Recorded arXiv version: v1; Recorded version updated: 2026-07-16T00:35:25Z.", "title": "Tactile: Giving Computer-Using Agents Hands and Feet", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.17050v1", "archive": "arXiv", "archive_location": "2607.17050", "author": [ { "literal": "Yaohan Yang" }, { "literal": "Minglei Shi" }, { "literal": "Borui Zhang" }, { "literal": "Jie Zhou" }, { "literal": "Jiwen Lu" } ], "citation-key": "arxiv_2607_17050", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6488798626abbec89ca739a0c5d188ae440484eb59b3be11f2b319ce742b22ec", "submitted_at": "2026-07-19T03:29:08Z", "updated_at": "2026-07-19T03:29:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.17050v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.17050", "catalog_id": "arxiv:2607.17050", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-19", "identifiers": { "arxiv": "2607.17050" }, "record_updated_at": "2026-07-19" }, "genre": "preprint", "id": "arxiv:2607.17050", "issued": { "date-parts": [ [ 2026, 7, 19 ] ] }, "note": "First published: 2026-07-19; Recorded arXiv version: v1; Recorded version updated: 2026-07-19T03:29:08Z.", "title": "EvoGUI: An Evolution-Aware Benchmark for GUI State-Transition Understanding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.18659v1", "archive": "arXiv", "archive_location": "2607.18659", "author": [ { "literal": "Behzad Ousat" }, { "literal": "Nikita Turkmen" }, { "literal": "Lalchandra Rampersaud" }, { "literal": "Dillan Bailey" }, { "literal": "Amin Kharraz" } ], "citation-key": "arxiv_2607_18659", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e382b15a9d25e99aa187a72c7d2a33bbeca82b2b347ba580ca0a1d73a8e01d19", "submitted_at": "2026-07-21T03:02:57Z", "updated_at": "2026-07-21T03:02:57Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.18659v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.18659", "catalog_id": "arxiv:2607.18659", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-21", "identifiers": { "arxiv": "2607.18659" }, "record_updated_at": "2026-07-21" }, "genre": "preprint", "id": "arxiv:2607.18659", "issued": { "date-parts": [ [ 2026, 7, 21 ] ] }, "note": "First published: 2026-07-21; Recorded arXiv version: v1; Recorded version updated: 2026-07-21T03:02:57Z.", "title": "Broken Gates: Re-evaluating Web Bot Defenses in the Age of LLM Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.22689v1", "archive": "arXiv", "archive_location": "2607.22689", "author": [ { "literal": "Zedong Yu" }, { "literal": "Qianxing Li" }, { "literal": "Zhi Gao" }, { "literal": "Liuyu Xiang" }, { "literal": "Chenrui Shi" }, { "literal": "Yang Liu" }, { "literal": "Huiming Wu" }, { "literal": "Yujie Wei" }, { "literal": "Yuhao Fei" }, { "literal": "Yubo Fu" }, { "literal": "Zhaofeng He" } ], "citation-key": "arxiv_2607_22689", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "078ff9a56b769923d15a98581685a3e2f47aac7d898ea731999cabaec3918fa7", "submitted_at": "2026-07-17T12:48:03Z", "updated_at": "2026-07-17T12:48:03Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.22689v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.22689", "catalog_id": "arxiv:2607.22689", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-17", "identifiers": { "arxiv": "2607.22689" }, "record_updated_at": "2026-07-17" }, "genre": "preprint", "id": "arxiv:2607.22689", "issued": { "date-parts": [ [ 2026, 7, 17 ] ] }, "note": "First published: 2026-07-17; Recorded arXiv version: v1; Recorded version updated: 2026-07-17T12:48:03Z.", "title": "Beyond Sequential Interaction: Benchmarking Parallel Execution and Coordination for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.23263v2", "archive": "arXiv", "archive_location": "2607.23263", "author": [ { "literal": "Yang Wan" }, { "literal": "Zhenhao Zhang" }, { "literal": "Jierui Wang" }, { "literal": "Xiangxin Zhou" }, { "literal": "Linchao Zhu" } ], "citation-key": "arxiv_2607_23263", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2de35aa3c7f4457d43d485eb136d8b380ea61870cc7253e807880ae731c51316", "submitted_at": "2026-07-25T16:00:45Z", "updated_at": "2026-09-26T12:15:44Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2607.23263v2" }, "canonical_source_url": "https://arxiv.org/abs/2607.23263", "catalog_id": "arxiv:2607.23263", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-25", "identifiers": { "arxiv": "2607.23263" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2607.23263", "issued": { "date-parts": [ [ 2026, 7, 25 ] ] }, "note": "First published: 2026-07-25; Recorded arXiv version: v2; Recorded version updated: 2026-09-26T12:15:44Z.", "title": "SeekJudge: A Practical Reward Framework for Reinforcement Learning in Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2607.25904v2", "archive": "arXiv", "archive_location": "2607.25904", "author": [ { "literal": "Chenrui Shi" }, { "literal": "Yuwei Wu" }, { "literal": "Yang Liu" }, { "literal": "Ruining Feng" }, { "literal": "Zirui Shang" }, { "literal": "Zhi Gao" }, { "literal": "Lifeng Fan" }, { "literal": "Che Sun" } ], "citation-key": "arxiv_2607_25904", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d1688a528d35a8213a8027385b1b1327536567f6ec307aa6562eea543ee71d7f", "submitted_at": "2026-07-28T16:01:38Z", "updated_at": "2026-07-29T05:13:57Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2607.25904v2" }, "canonical_source_url": "https://arxiv.org/abs/2607.25904", "catalog_id": "arxiv:2607.25904", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-28", "identifiers": { "arxiv": "2607.25904" }, "record_updated_at": "2026-07-29" }, "genre": "preprint", "id": "arxiv:2607.25904", "issued": { "date-parts": [ [ 2026, 7, 28 ] ] }, "note": "First published: 2026-07-28; Recorded arXiv version: v2; Recorded version updated: 2026-07-29T05:13:57Z.", "title": "Interactive Reward Agent: GUI Task Evaluation via Environment-State Verification", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2607.26041v2", "archive": "arXiv", "archive_location": "2607.26041", "author": [ { "literal": "Abhishek Pillai" }, { "literal": "Samir Kumar Nayak" }, { "literal": "Yuan Chen" } ], "citation-key": "arxiv_2607_26041", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "352c892a52692301e14125e126a52508feb39ef7cd034d7cb0c48a27fb54b8ac", "submitted_at": "2026-07-28T17:49:51Z", "updated_at": "2026-07-29T19:07:50Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2607.26041v2" }, "canonical_source_url": "https://arxiv.org/abs/2607.26041", "catalog_id": "arxiv:2607.26041", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-28", "identifiers": { "arxiv": "2607.26041" }, "record_updated_at": "2026-07-29" }, "genre": "preprint", "id": "arxiv:2607.26041", "issued": { "date-parts": [ [ 2026, 7, 28 ] ] }, "note": "First published: 2026-07-28; Recorded arXiv version: v2; Recorded version updated: 2026-07-29T19:07:50Z.", "title": "Desktop-Delta Bench: Do Computer-Use Models Understand Desktop GUI Transitions?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2607.28367v1", "archive": "arXiv", "archive_location": "2607.28367", "author": [ { "literal": "Zihan Dong" }, { "literal": "Zhiyuan Ma" }, { "literal": "Zekun Wang" }, { "literal": "Yunqing Li" }, { "literal": "Zirou Liu" }, { "literal": "Ruixuan Deng" }, { "literal": "Qishi Zhan" }, { "literal": "Rui Qian" } ], "citation-key": "arxiv_2607_28367", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d75f6c942d6854a259556b7ef1060fa0f142755162da329580c6177cd1adef15", "submitted_at": "2026-07-30T15:29:42Z", "updated_at": "2026-07-30T15:29:42Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.28367v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.28367", "catalog_id": "arxiv:2607.28367", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-30", "identifiers": { "arxiv": "2607.28367" }, "record_updated_at": "2026-07-30" }, "genre": "preprint", "id": "arxiv:2607.28367", "issued": { "date-parts": [ [ 2026, 7, 30 ] ] }, "note": "First published: 2026-07-30; Recorded arXiv version: v1; Recorded version updated: 2026-07-30T15:29:42Z.", "title": "How Benchmarks Mis-Score Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.28573v1", "archive": "arXiv", "archive_location": "2607.28573", "author": [ { "literal": "Woongkyu Lee" }, { "literal": "Jungwook Choi" } ], "citation-key": "arxiv_2607_28573", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bacd1c5df5bf3163c83fae546ae4ced36996f33115568dfbc9e44b17fe507a31", "submitted_at": "2026-07-30T17:36:36Z", "updated_at": "2026-07-30T17:36:36Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2607.28573v1" }, "canonical_source_url": "https://arxiv.org/abs/2607.28573", "catalog_id": "arxiv:2607.28573", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-30", "identifiers": { "arxiv": "2607.28573" }, "record_updated_at": "2026-07-30" }, "genre": "preprint", "id": "arxiv:2607.28573", "issued": { "date-parts": [ [ 2026, 7, 30 ] ] }, "note": "First published: 2026-07-30; Recorded arXiv version: v1; Recorded version updated: 2026-07-30T17:36:36Z.", "title": "Rethinking Inference-Time Scaling in Local Computer-Use Agents: Failure Modes and Compute Tradeoffs", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2607.28609v2", "archive": "arXiv", "archive_location": "2607.28609", "author": [ { "literal": "Qiushi Sun" }, { "literal": "Kanzhi Cheng" }, { "literal": "Yian Wang" }, { "literal": "Bowen Yang" }, { "literal": "Hang Yan" }, { "literal": "Liheng Chen" }, { "literal": "Fangzhi Xu" }, { "literal": "Zichen Ding" }, { "literal": "Nuo Chen" }, { "literal": "Jialin Cao" }, { "literal": "Xingdong Gong" }, { "literal": "Zehao Li" }, { "literal": "Kaiming Jin" }, { "literal": "Xinfeng Yuan" }, { "literal": "Zhoumianze Liu" }, { "literal": "Jingyang Gong" }, { "literal": "Zhangyue Yin" }, { "literal": "Jiahui Gao" }, { "literal": "Zhiyong Wu" }, { "literal": "Tianbao Xie" }, { "literal": "Jianbing Zhang" }, { "literal": "Ben Kao" }, { "literal": "Lingpeng Kong" } ], "citation-key": "arxiv_2607_28609", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e3e82df7abffbd0c8f7e969ed8edb77423263a2b1d826a3e72529b0a51903d58", "submitted_at": "2026-07-30T17:57:41Z", "updated_at": "2026-08-06T17:55:59Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2607.28609v2" }, "canonical_source_url": "https://arxiv.org/abs/2607.28609", "catalog_id": "arxiv:2607.28609", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-30", "identifiers": { "arxiv": "2607.28609" }, "record_updated_at": "2026-08-06" }, "genre": "preprint", "id": "arxiv:2607.28609", "issued": { "date-parts": [ [ 2026, 7, 30 ] ] }, "note": "First published: 2026-07-30; Recorded arXiv version: v2; Recorded version updated: 2026-08-06T17:55:59Z.", "title": "OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2608.02352v1", "archive": "arXiv", "archive_location": "2608.02352", "author": [ { "literal": "Dunjie Lu" }, { "literal": "Shuai Bai" }, { "literal": "Tianyi Bai" }, { "literal": "Sicheng Fan" }, { "literal": "Chang Gao" }, { "literal": "Jian Guan" }, { "literal": "Feng Hu" }, { "literal": "Mianqiu Huang" }, { "literal": "Xingyang Huang" }, { "literal": "Yizhen Jiang" }, { "literal": "Yuheng Jing" }, { "literal": "Dehui Kong" }, { "literal": "Ning Li" }, { "literal": "Dayiheng Liu" }, { "literal": "Shixuan Liu" }, { "literal": "Zheng Liu" }, { "literal": "Que Shen" }, { "literal": "Bowen Wang" }, { "literal": "Junli Wang" }, { "literal": "Chencan Wu" }, { "literal": "Rui Xie" }, { "literal": "Tianbao Xie" }, { "literal": "Zhihui Xie" }, { "literal": "Haiyang Xu" }, { "literal": "An Yang" }, { "literal": "Tao Yu" }, { "literal": "Wenzhen Yuan" }, { "literal": "Xi Zhang" }, { "literal": "Zhenru Zhang" }, { "literal": "Mingkang Zhu" }, { "literal": "Zhaoqing Zhu" }, { "literal": "Yizhong Cao" }, { "literal": "Kai Dang" }, { "literal": "Binyuan Hui" }, { "literal": "Kaixin Li" }, { "literal": "Junyang Lin" }, { "literal": "Haiquan Wang" }, { "literal": "Zekun Wang" }, { "literal": "Yiheng Xu" }, { "literal": "Fan Yan" }, { "literal": "Mengqi Yuan" }, { "literal": "Danyang Zhang" }, { "literal": "Jiajun Zhang" }, { "literal": "Zhipeng Zhang" }, { "literal": "Fan Zhou" }, { "literal": "Fan Zhou" } ], "citation-key": "arxiv_2608_02352", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "26a61ccfdddb5cdc2da2b14d29ae8d4098544fea4e9458126f2a49c28418c4f0", "submitted_at": "2026-08-03T15:04:20Z", "updated_at": "2026-08-03T15:04:20Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.02352v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.02352", "catalog_id": "arxiv:2608.02352", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-03", "identifiers": { "arxiv": "2608.02352" }, "record_updated_at": "2026-08-03" }, "genre": "preprint", "id": "arxiv:2608.02352", "issued": { "date-parts": [ [ 2026, 8, 3 ] ] }, "note": "First published: 2026-08-03; Recorded arXiv version: v1; Recorded version updated: 2026-08-03T15:04:20Z.", "title": "Qwen-CUA: Native Computer Use for (almost) Everything", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.02643v2", "archive": "arXiv", "archive_location": "2608.02643", "author": [ { "literal": "Weijia Zhang" }, { "literal": "Kunlun Zhu" }, { "literal": "Zeyi Liu" }, { "literal": "Yinting Chen" }, { "literal": "Tianyi Ma" }, { "literal": "Jiateng Liu" }, { "literal": "Jiaxun Zhang" }, { "literal": "Bingxuan Li" }, { "literal": "Xiangru Tang" }, { "literal": "Heng Ji" }, { "literal": "Jiaxuan You" } ], "citation-key": "arxiv_2608_02643", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a11e065bef4cf860c36ca2f682801ca21b552fbd40bd1b6f5d9ae39b56e7d77a", "submitted_at": "2026-07-31T13:59:45Z", "updated_at": "2026-09-06T04:43:23Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2608.02643v2" }, "canonical_source_url": "https://arxiv.org/abs/2608.02643", "catalog_id": "arxiv:2608.02643", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-31", "identifiers": { "arxiv": "2608.02643" }, "record_updated_at": "2026-09-06" }, "genre": "preprint", "id": "arxiv:2608.02643", "issued": { "date-parts": [ [ 2026, 7, 31 ] ] }, "note": "First published: 2026-07-31; Recorded arXiv version: v2; Recorded version updated: 2026-09-06T04:43:23Z.", "title": "CUADebug: Diagnosing and Repairing Computer-Use Agent Failures", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2608.03270v1", "archive": "arXiv", "archive_location": "2608.03270", "author": [ { "literal": "Zichuan Fu" }, { "literal": "Shirong Wang" }, { "literal": "Wenlin Zhang" }, { "literal": "Guojing Li" }, { "literal": "Yimin Deng" }, { "literal": "Jingtong Gao" }, { "literal": "Junjia Qi" }, { "literal": "Hanyu Yan" }, { "literal": "Yefeng Zheng" }, { "literal": "Xiaopeng Li" }, { "literal": "Wanyu Wang" }, { "literal": "Xian Wu" }, { "literal": "Xiangyu Zhao" } ], "citation-key": "arxiv_2608_03270", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fb64bb5faa37e68b35a2398ad896fc02181a4d2749a3067391e6b3bf023a0f48", "submitted_at": "2026-08-04T07:47:37Z", "updated_at": "2026-08-04T07:47:37Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.03270v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.03270", "catalog_id": "arxiv:2608.03270", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-04", "identifiers": { "arxiv": "2608.03270" }, "record_updated_at": "2026-08-04" }, "genre": "preprint", "id": "arxiv:2608.03270", "issued": { "date-parts": [ [ 2026, 8, 4 ] ] }, "note": "First published: 2026-08-04; Recorded arXiv version: v1; Recorded version updated: 2026-08-04T07:47:37Z.", "title": "GUI-Lens: Coarse-to-Fine Cropping for GUI Grounding with General-Purpose VLMs", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.03327v2", "archive": "arXiv", "archive_location": "2608.03327", "author": [ { "literal": "Siqi Fan" }, { "literal": "Minghao Li" }, { "literal": "Xiaoqian Ma" }, { "literal": "Wenhui Tan" }, { "literal": "Xiusheng Huang" }, { "literal": "Juntong Wu" }, { "literal": "Liujie Zhang" }, { "literal": "Shuo Shang" }, { "literal": "Weihang Chen" } ], "citation-key": "arxiv_2608_03327", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2c8692712194d0ff822091d13324e6bb540339103a824f1030f7a971699d0a0b", "submitted_at": "2026-08-04T08:35:51Z", "updated_at": "2026-08-06T07:26:24Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2608.03327v2" }, "canonical_source_url": "https://arxiv.org/abs/2608.03327", "catalog_id": "arxiv:2608.03327", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-04", "identifiers": { "arxiv": "2608.03327" }, "record_updated_at": "2026-08-06" }, "genre": "preprint", "id": "arxiv:2608.03327", "issued": { "date-parts": [ [ 2026, 8, 4 ] ] }, "note": "First published: 2026-08-04; Recorded arXiv version: v2; Recorded version updated: 2026-08-06T07:26:24Z.", "title": "Screenshots or Tools? Eliciting Tool Use and Managing Multimodal Context in Hybrid GUI-MCP Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2608.03740v2", "archive": "arXiv", "archive_location": "2608.03740", "author": [ { "literal": "Yu Ran" }, { "literal": "Wentao Zhao" } ], "citation-key": "arxiv_2608_03740", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "152e21cd4f1f290aafadfdcfaa10e15a03378404540a46b5f1a5464a2d432d23", "submitted_at": "2026-08-04T14:34:41Z", "updated_at": "2026-09-28T13:07:55Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2608.03740v2" }, "canonical_source_url": "https://arxiv.org/abs/2608.03740", "catalog_id": "arxiv:2608.03740", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-04", "identifiers": { "arxiv": "2608.03740" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2608.03740", "issued": { "date-parts": [ [ 2026, 8, 4 ] ] }, "note": "First published: 2026-08-04; Recorded arXiv version: v2; Recorded version updated: 2026-09-28T13:07:55Z.", "title": "MissClick: Execution-Aware Adversarial Attacks on Coordinate Generation in GUI Grounding Models", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2608.04741v1", "archive": "arXiv", "archive_location": "2608.04741", "author": [ { "literal": "Longtao Guo" }, { "literal": "Zelin Zhang" }, { "literal": "Kaifeng Huang" }, { "literal": "Yang Shi" } ], "citation-key": "arxiv_2608_04741", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "30e5ce3d86881041491ae507839b2f3d302691aea1d2fdfc92cfc00f35efefe7", "submitted_at": "2026-08-05T12:08:16Z", "updated_at": "2026-08-05T12:08:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.04741v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.04741", "catalog_id": "arxiv:2608.04741", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-05", "identifiers": { "arxiv": "2608.04741" }, "record_updated_at": "2026-08-05" }, "genre": "preprint", "id": "arxiv:2608.04741", "issued": { "date-parts": [ [ 2026, 8, 5 ] ] }, "note": "First published: 2026-08-05; Recorded arXiv version: v1; Recorded version updated: 2026-08-05T12:08:16Z.", "title": "LoginTrap: Uncovering Task-Agnostic Phishing-Style Indirect Prompt Injection Attacks against LLM-based Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.04755v1", "archive": "arXiv", "archive_location": "2608.04755", "author": [ { "literal": "Dongsheng Chen" }, { "literal": "Yuxuan Li" }, { "literal": "Guanhua Chen" }, { "literal": "Jiaxin Zhang" }, { "literal": "Xiangyu Zhao" }, { "literal": "Lei Ma" }, { "literal": "Xin Yao" }, { "literal": "Xuetao Wei" } ], "citation-key": "arxiv_2608_04755", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5f8b84135f72558eea1fae428df7cb30f62232631de49e690e73010aeca22d4c", "submitted_at": "2026-08-05T12:23:16Z", "updated_at": "2026-08-05T12:23:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.04755v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.04755", "catalog_id": "arxiv:2608.04755", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-05", "identifiers": { "arxiv": "2608.04755" }, "record_updated_at": "2026-08-05" }, "genre": "preprint", "id": "arxiv:2608.04755", "issued": { "date-parts": [ [ 2026, 8, 5 ] ] }, "note": "First published: 2026-08-05; Recorded arXiv version: v1; Recorded version updated: 2026-08-05T12:23:16Z.", "title": "\"Allow\" to Achieve, Over-Privileged Inadvertently: The Unintended Cost of Task-Completion-Driven Pop-up Decisions in Mobile GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.05587v1", "archive": "arXiv", "archive_location": "2608.05587", "author": [ { "literal": "Linqiang Guo" }, { "literal": "Wei Liu" }, { "literal": "Li Gu" }, { "literal": "Yang Wang" }, { "literal": "Tse-Hsun" }, { "literal": "Chen" } ], "citation-key": "arxiv_2608_05587", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "dc7934927c82a43e6ae6a246ade1d50c4551cc49d263ca67cfb32906cbec9ea4", "submitted_at": "2026-08-06T04:17:27Z", "updated_at": "2026-08-06T04:17:27Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.05587v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.05587", "catalog_id": "arxiv:2608.05587", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-06", "identifiers": { "arxiv": "2608.05587" }, "record_updated_at": "2026-08-06" }, "genre": "preprint", "id": "arxiv:2608.05587", "issued": { "date-parts": [ [ 2026, 8, 6 ] ] }, "note": "First published: 2026-08-06; Recorded arXiv version: v1; Recorded version updated: 2026-08-06T04:17:27Z.", "title": "StepReflect: Structured UI Transition Reflection for Mobile GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.06171v1", "archive": "arXiv", "archive_location": "2608.06171", "author": [ { "literal": "Jiaming Wei" }, { "literal": "Zekun Wu" }, { "literal": "Adriano Koshiyama" }, { "literal": "Maria Perez-Ortiz" } ], "citation-key": "arxiv_2608_06171", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9055ec9c70777fc94179983d82bf841bac12332ebed7f49342883c8d842d0fbc", "submitted_at": "2026-08-06T15:37:04Z", "updated_at": "2026-08-06T15:37:04Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.06171v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.06171", "catalog_id": "arxiv:2608.06171", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-06", "identifiers": { "arxiv": "2608.06171" }, "record_updated_at": "2026-08-06" }, "genre": "preprint", "id": "arxiv:2608.06171", "issued": { "date-parts": [ [ 2026, 8, 6 ] ] }, "note": "First published: 2026-08-06; Recorded arXiv version: v1; Recorded version updated: 2026-08-06T15:37:04Z.", "title": "Routing Is Least Learnable Where It Is Most Valuable: Bounds on Representation Routing for Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.07775v1", "archive": "arXiv", "archive_location": "2608.07775", "author": [ { "literal": "Xiaoou Liu" }, { "literal": "Longchao Da" }, { "literal": "Hanyang Chen" }, { "literal": "Yuan Ling" }, { "literal": "Hua Wei" } ], "citation-key": "arxiv_2608_07775", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "4d99685a75d2023db5c9ad943e103ec779eebd037fe7a113848d8a31e08fd68b", "submitted_at": "2026-08-07T21:47:43Z", "updated_at": "2026-08-07T21:47:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.07775v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.07775", "catalog_id": "arxiv:2608.07775", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-07", "identifiers": { "arxiv": "2608.07775" }, "record_updated_at": "2026-08-07" }, "genre": "preprint", "id": "arxiv:2608.07775", "issued": { "date-parts": [ [ 2026, 8, 7 ] ] }, "note": "First published: 2026-08-07; Recorded arXiv version: v1; Recorded version updated: 2026-08-07T21:47:43Z.", "title": "AndroidReality: How Far Are Mobile Agents from the Real World?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.08392v1", "archive": "arXiv", "archive_location": "2608.08392", "author": [ { "literal": "Zejun Xu" }, { "literal": "Taiyi Chen" }, { "literal": "Jin Li" }, { "literal": "Yongtong Gu" }, { "literal": "Qi Cheng" }, { "literal": "Aixuan Lv" }, { "literal": "Shuai Zhu" }, { "literal": "Pengfei Zhu" }, { "literal": "Kaichen Yang" }, { "literal": "Boyu Sun" }, { "literal": "Yixian Yang" }, { "literal": "Mulong Xie" }, { "literal": "Xin Liu" }, { "literal": "Dagang Li" }, { "literal": "Xiaoteng Ma" }, { "literal": "Hongru Wang" } ], "citation-key": "arxiv_2608_08392", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d0e469cd7d84080e11df46d826460bea79eae71b8fa5786cacd96aac02b8b0a8", "submitted_at": "2026-08-09T01:08:19Z", "updated_at": "2026-08-09T01:08:19Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.08392v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.08392", "catalog_id": "arxiv:2608.08392", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-09", "identifiers": { "arxiv": "2608.08392" }, "record_updated_at": "2026-08-09" }, "genre": "preprint", "id": "arxiv:2608.08392", "issued": { "date-parts": [ [ 2026, 8, 9 ] ] }, "note": "First published: 2026-08-09; Recorded arXiv version: v1; Recorded version updated: 2026-08-09T01:08:19Z.", "title": "CAP: A Scalable Benchmark for Evaluating Cross-Site Browser Agents with Complex Actions and Perception", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.09654v1", "archive": "arXiv", "archive_location": "2608.09654", "author": [ { "literal": "Yuke Li" }, { "literal": "Xuehan Hou" } ], "citation-key": "arxiv_2608_09654", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f05f128cccb514987b7defb52c8f43b45d917e0e0f9570db4e75c4fe23b38533", "submitted_at": "2026-08-10T14:29:07Z", "updated_at": "2026-08-10T14:29:07Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.09654v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.09654", "catalog_id": "arxiv:2608.09654", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-10", "identifiers": { "arxiv": "2608.09654" }, "record_updated_at": "2026-08-10" }, "genre": "preprint", "id": "arxiv:2608.09654", "issued": { "date-parts": [ [ 2026, 8, 10 ] ] }, "note": "First published: 2026-08-10; Recorded arXiv version: v1; Recorded version updated: 2026-08-10T14:29:07Z.", "title": "Hallucination-Free GUI Grounding via Regression-Free Layout-Aware Matching", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.09944v1", "archive": "arXiv", "archive_location": "2608.09944", "author": [ { "literal": "Santosh Patapati" } ], "citation-key": "arxiv_2608_09944", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d5ccb0d41798a1a3236e6f9b4361814cccdd40facf9b43bf0bf52932337484dc", "submitted_at": "2026-07-01T16:24:07Z", "updated_at": "2026-07-01T16:24:07Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.09944v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.09944", "catalog_id": "arxiv:2608.09944", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-07-01", "identifiers": { "arxiv": "2608.09944" }, "record_updated_at": "2026-07-01" }, "genre": "preprint", "id": "arxiv:2608.09944", "issued": { "date-parts": [ [ 2026, 7, 1 ] ] }, "note": "First published: 2026-07-01; Recorded arXiv version: v1; Recorded version updated: 2026-07-01T16:24:07Z.", "title": "Navigation Alone Is Not Enough: Evaluating Explanatory Assistive UI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.10366v1", "archive": "arXiv", "archive_location": "2608.10366", "author": [ { "literal": "Mizanur Rahman" }, { "literal": "Mohammed Saidul Islam" }, { "literal": "Ridwan Mahbub" }, { "literal": "Md Tahmid Rahman Laskar" }, { "literal": "Shafiq Joty" }, { "literal": "Enamul Hoque Prince" } ], "citation-key": "arxiv_2608_10366", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e1089e57e0f0146f29a4a411c5c6f72da3c9f45be223fbbc5c9630315be6a680", "submitted_at": "2026-08-11T01:45:56Z", "updated_at": "2026-08-11T01:45:56Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.10366v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.10366", "catalog_id": "arxiv:2608.10366", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-11", "identifiers": { "arxiv": "2608.10366" }, "record_updated_at": "2026-08-11" }, "genre": "preprint", "id": "arxiv:2608.10366", "issued": { "date-parts": [ [ 2026, 8, 11 ] ] }, "note": "First published: 2026-08-11; Recorded arXiv version: v1; Recorded version updated: 2026-08-11T01:45:56Z.", "title": "DSAgentBench: Can Agents Automate End-to-End Data-Science Workflows in Real Computer Environments?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.11434v1", "archive": "arXiv", "archive_location": "2608.11434", "author": [ { "literal": "Ziqiang Wang" }, { "literal": "Li Gu" }, { "literal": "Zhixiang Chi" }, { "literal": "Zhi Liu" }, { "literal": "Seyed Mehdi Ayyoubzadeh" }, { "literal": "Yuanhao Yu" }, { "literal": "Yang Wang" } ], "citation-key": "arxiv_2608_11434", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a2606ae96ca3ea99b5c7278485601f9bde3b64eefdd3bfecc2bbfc678c39b4fc", "submitted_at": "2026-08-11T21:00:46Z", "updated_at": "2026-08-11T21:00:46Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.11434v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.11434", "catalog_id": "arxiv:2608.11434", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-11", "identifiers": { "arxiv": "2608.11434" }, "record_updated_at": "2026-08-11" }, "genre": "preprint", "id": "arxiv:2608.11434", "issued": { "date-parts": [ [ 2026, 8, 11 ] ] }, "note": "First published: 2026-08-11; Recorded arXiv version: v1; Recorded version updated: 2026-08-11T21:00:46Z.", "title": "Benchmarking LLM Judges for Mobile Agent Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.14131v1", "archive": "arXiv", "archive_location": "2608.14131", "author": [ { "literal": "Thilo Reintjes" }, { "literal": "Sivajeet Chand" }, { "literal": "Derui Zhu" }, { "literal": "Sushant Kumar Pandey" }, { "literal": "Alexander Pretschner" } ], "citation-key": "arxiv_2608_14131", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f8a6bd69d421bdfb9a21c22e715a59d0afb0b0cab75510f28ff569903ecffae5", "submitted_at": "2026-08-14T09:36:33Z", "updated_at": "2026-08-14T09:36:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.14131v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.14131", "catalog_id": "arxiv:2608.14131", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-14", "identifiers": { "arxiv": "2608.14131" }, "record_updated_at": "2026-08-14" }, "genre": "preprint", "id": "arxiv:2608.14131", "issued": { "date-parts": [ [ 2026, 8, 14 ] ] }, "note": "First published: 2026-08-14; Recorded arXiv version: v1; Recorded version updated: 2026-08-14T09:36:33Z.", "title": "LegacyWorld: Atomicity-Aware Evaluation of GUI Agents for Legacy Workflows", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.14132v1", "archive": "arXiv", "archive_location": "2608.14132", "author": [ { "literal": "Xiaokai Yan" }, { "literal": "Jingtao Ding" }, { "literal": "Yong Li" }, { "literal": "Zhiwen Yu" } ], "citation-key": "arxiv_2608_14132", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e2660e01921a78bed7280b34f35a1b587e62339c2e280d6294bbb6d6e666786e", "submitted_at": "2026-08-14T09:36:59Z", "updated_at": "2026-08-14T09:36:59Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.14132v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.14132", "catalog_id": "arxiv:2608.14132", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-14", "identifiers": { "arxiv": "2608.14132" }, "record_updated_at": "2026-08-14" }, "genre": "preprint", "id": "arxiv:2608.14132", "issued": { "date-parts": [ [ 2026, 8, 14 ] ] }, "note": "First published: 2026-08-14; Recorded arXiv version: v1; Recorded version updated: 2026-08-14T09:36:59Z.", "title": "Act2Intention: A Benchmark For Developing Active Mobile Agents Through Inferring User Intention from GUI Actions", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.17659v1", "archive": "arXiv", "archive_location": "2608.17659", "author": [ { "literal": "Sujin Chen" }, { "literal": "Lijun Li" }, { "literal": "Tianyi Du" }, { "literal": "Jing Shao" } ], "citation-key": "arxiv_2608_17659", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c49a67b38314152b83e793b9ade7ce4102a44aa27c331504a8758b94df906c67", "submitted_at": "2026-08-18T11:33:22Z", "updated_at": "2026-08-18T11:33:22Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.17659v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.17659", "catalog_id": "arxiv:2608.17659", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-18", "identifiers": { "arxiv": "2608.17659" }, "record_updated_at": "2026-08-18" }, "genre": "preprint", "id": "arxiv:2608.17659", "issued": { "date-parts": [ [ 2026, 8, 18 ] ] }, "note": "First published: 2026-08-18; Recorded arXiv version: v1; Recorded version updated: 2026-08-18T11:33:22Z.", "title": "MobileWorldSafety: Benchmarking GUI Agent Safety Against Environmental Injection Attacks in Android Apps", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.18307v1", "archive": "arXiv", "archive_location": "2608.18307", "author": [ { "literal": "Tianchen Guan" }, { "literal": "Xinlei Lin" }, { "literal": "Royce Cheng-Yue" }, { "literal": "Xiangjun Wang" }, { "literal": "Shuyan Zhou" } ], "citation-key": "arxiv_2608_18307", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "de4889c021f35c169092e17801fec414cce21b37c74a4ff72af6a6c67a1a0a25", "submitted_at": "2026-08-18T20:38:26Z", "updated_at": "2026-08-18T20:38:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.18307v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.18307", "catalog_id": "arxiv:2608.18307", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-18", "identifiers": { "arxiv": "2608.18307" }, "record_updated_at": "2026-08-18" }, "genre": "preprint", "id": "arxiv:2608.18307", "issued": { "date-parts": [ [ 2026, 8, 18 ] ] }, "note": "First published: 2026-08-18; Recorded arXiv version: v1; Recorded version updated: 2026-08-18T20:38:26Z.", "title": "ComponentBench: Diagnosing Component-Level Failures in Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.21794v1", "archive": "arXiv", "archive_location": "2608.21794", "author": [ { "literal": "Qijia Chen" }, { "literal": "Giulio Jacucci" } ], "citation-key": "arxiv_2608_21794", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "cabd094270d01b20df7b029be29c5d460d10beb84e43e3e783f4f11997b74114", "submitted_at": "2026-08-22T06:30:03Z", "updated_at": "2026-08-22T06:30:03Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.21794v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.21794", "catalog_id": "arxiv:2608.21794", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-22", "identifiers": { "arxiv": "2608.21794" }, "record_updated_at": "2026-08-22" }, "genre": "preprint", "id": "arxiv:2608.21794", "issued": { "date-parts": [ [ 2026, 8, 22 ] ] }, "note": "First published: 2026-08-22; Recorded arXiv version: v1; Recorded version updated: 2026-08-22T06:30:03Z.", "title": "Lexical Coupling in GUI Element Grounding: Sentence Embeddings Track Labels across Mobile and Web", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.21832v2", "archive": "arXiv", "archive_location": "2608.21832", "author": [ { "literal": "Md Abrar Jahin" }, { "literal": "Md Rizwan Parvez" } ], "citation-key": "arxiv_2608_21832", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "970441704314fabffc46cb4a340dd6e0bde0e84ab9b7fcd2db2c9a8be9dc61d8", "submitted_at": "2026-08-22T08:04:59Z", "updated_at": "2026-08-27T07:03:01Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2608.21832v2" }, "canonical_source_url": "https://arxiv.org/abs/2608.21832", "catalog_id": "arxiv:2608.21832", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-22", "identifiers": { "arxiv": "2608.21832" }, "record_updated_at": "2026-08-27" }, "genre": "preprint", "id": "arxiv:2608.21832", "issued": { "date-parts": [ [ 2026, 8, 22 ] ] }, "note": "First published: 2026-08-22; Recorded arXiv version: v2; Recorded version updated: 2026-08-27T07:03:01Z.", "title": "GUI-Primitives: Diagnosing Spatial Reasoning Failures in Vision-Language GUI Grounding", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2608.24099v1", "archive": "arXiv", "archive_location": "2608.24099", "author": [ { "literal": "Guo Gan" }, { "literal": "Yilun Zhao" }, { "literal": "Cong Chen" }, { "literal": "Jinbiao Wei" }, { "literal": "Tingyu Song" }, { "literal": "Zheyuan Yang" }, { "literal": "Lin Fu" }, { "literal": "Hong Zhou" } ], "citation-key": "arxiv_2608_24099", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "51901009de5898c52a1b9d4958b698808a8f8b850d19afa32dd7a3f32d42a851", "submitted_at": "2026-08-25T05:57:25Z", "updated_at": "2026-08-25T05:57:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.24099v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.24099", "catalog_id": "arxiv:2608.24099", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-25", "identifiers": { "arxiv": "2608.24099" }, "record_updated_at": "2026-08-25" }, "genre": "preprint", "id": "arxiv:2608.24099", "issued": { "date-parts": [ [ 2026, 8, 25 ] ] }, "note": "First published: 2026-08-25; Recorded arXiv version: v1; Recorded version updated: 2026-08-25T05:57:25Z.", "title": "Are Android GUI Agents Robust Against Runtime Anomalies? AnTrap: Evaluating Agents in Dynamic Adversarial Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.24174v1", "archive": "arXiv", "archive_location": "2608.24174", "author": [ { "literal": "Tao Xiong" }, { "literal": "Xavier Hu" }, { "literal": "Wenkai Wang" }, { "literal": "Qinzhuo Wu" }, { "literal": "Changqiao Wu" }, { "literal": "Pengzhi Gao" }, { "literal": "Wei Liu" }, { "literal": "Jian Luan" }, { "literal": "Shengyu Zhang" } ], "citation-key": "arxiv_2608_24174", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "76cd09f70bfec0d2a49cddae8adcf32ca37fb39bcade7c1d36d5f3dc3fa6bfa7", "submitted_at": "2026-08-25T07:40:12Z", "updated_at": "2026-08-25T07:40:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.24174v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.24174", "catalog_id": "arxiv:2608.24174", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-25", "identifiers": { "arxiv": "2608.24174" }, "record_updated_at": "2026-08-25" }, "genre": "preprint", "id": "arxiv:2608.24174", "issued": { "date-parts": [ [ 2026, 8, 25 ] ] }, "note": "First published: 2026-08-25; Recorded arXiv version: v1; Recorded version updated: 2026-08-25T07:40:12Z.", "title": "Task-Adaptive Rubrics for GUI Reward Modeling", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.26204v2", "archive": "arXiv", "archive_location": "2608.26204", "author": [ { "literal": "Joy Chen" }, { "literal": "Alejandro Castillejo Munoz" }, { "literal": "Pierluca D'Oro" }, { "literal": "Yuxuan Sun" }, { "literal": "Chloe Evans" }, { "literal": "Joseph Tighe" } ], "citation-key": "arxiv_2608_26204", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "50c26a7ab741ecb3cb855305ee4f7d5b5729296248d650c469600aeb73101781", "submitted_at": "2026-08-25T23:21:57Z", "updated_at": "2026-09-15T04:43:00Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2608.26204v2" }, "canonical_source_url": "https://arxiv.org/abs/2608.26204", "catalog_id": "arxiv:2608.26204", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-25", "identifiers": { "arxiv": "2608.26204" }, "record_updated_at": "2026-09-15" }, "genre": "preprint", "id": "arxiv:2608.26204", "issued": { "date-parts": [ [ 2026, 8, 25 ] ] }, "note": "First published: 2026-08-25; Recorded arXiv version: v2; Recorded version updated: 2026-09-15T04:43:00Z.", "title": "ADeptS-Bench: Measuring the Trustworthiness of Computer Use Agents Across Devices", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2608.26991v1", "archive": "arXiv", "archive_location": "2608.26991", "author": [ { "literal": "Rui Xie" }, { "literal": "Lu Chen" } ], "citation-key": "arxiv_2608_26991", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e1626d5190f145c668c9b7f363b7f84904c12daae485e63b858e440ded5ea6df", "submitted_at": "2026-08-27T11:38:19Z", "updated_at": "2026-08-27T11:38:19Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.26991v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.26991", "catalog_id": "arxiv:2608.26991", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-27", "identifiers": { "arxiv": "2608.26991" }, "record_updated_at": "2026-08-27" }, "genre": "preprint", "id": "arxiv:2608.26991", "issued": { "date-parts": [ [ 2026, 8, 27 ] ] }, "note": "First published: 2026-08-27; Recorded arXiv version: v1; Recorded version updated: 2026-08-27T11:38:19Z.", "title": "ASIL: Replacing Screenshot-and-Click with Structured State and Semantic Actions", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.27808v1", "archive": "arXiv", "archive_location": "2608.27808", "author": [ { "literal": "Divake Kumar" }, { "literal": "Sina Tayebati" }, { "literal": "Devashri Naik" }, { "literal": "Amanda Sofie Rios" }, { "literal": "Nilesh Ahuja" }, { "literal": "Omesh Tickoo" }, { "literal": "Ranganath Krishnan" }, { "literal": "Amit Ranjan Trivedi" } ], "citation-key": "arxiv_2608_27808", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "14864aae8f977d474ff3e7dd3e38951374862abfb8077395f2dc5a9e1cc5a489", "submitted_at": "2026-08-28T00:50:47Z", "updated_at": "2026-08-28T00:50:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.27808v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.27808", "catalog_id": "arxiv:2608.27808", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-28", "identifiers": { "arxiv": "2608.27808" }, "record_updated_at": "2026-08-28" }, "genre": "preprint", "id": "arxiv:2608.27808", "issued": { "date-parts": [ [ 2026, 8, 28 ] ] }, "note": "First published: 2026-08-28; Recorded arXiv version: v1; Recorded version updated: 2026-08-28T00:50:47Z.", "title": "CURA: Certified Runtime Alarms for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.30038v1", "archive": "arXiv", "archive_location": "2608.30038", "author": [ { "literal": "Mingshuo Wang" }, { "literal": "Hanqing Guo" }, { "literal": "Huining Li" }, { "literal": "Yuliang Fu" }, { "literal": "Jing Xu" }, { "literal": "Chenhan Xu" } ], "citation-key": "arxiv_2608_30038", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7b17f858c9d6992f6a8d8f21cc79d468b0f1c81fc44084045791d9c3acacd6d0", "submitted_at": "2026-08-30T20:55:21Z", "updated_at": "2026-08-30T20:55:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2608.30038v1" }, "canonical_source_url": "https://arxiv.org/abs/2608.30038", "catalog_id": "arxiv:2608.30038", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-30", "identifiers": { "arxiv": "2608.30038" }, "record_updated_at": "2026-08-30" }, "genre": "preprint", "id": "arxiv:2608.30038", "issued": { "date-parts": [ [ 2026, 8, 30 ] ] }, "note": "First published: 2026-08-30; Recorded arXiv version: v1; Recorded version updated: 2026-08-30T20:55:21Z.", "title": "ActReal: System-Level Mobile Agents Challenge Mobile Automation Detection", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2608.30207v2", "archive": "arXiv", "archive_location": "2608.30207", "author": [ { "literal": "Chen Xiong" }, { "literal": "Zhiyuan He" }, { "literal": "Pin-Yu Chen" }, { "literal": "Stjepan Picek" }, { "literal": "Tsung-Yi Ho" } ], "citation-key": "arxiv_2608_30207", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "87c3f58eebe2db437b1166ff8e83b336db4cde37684a7cd86f61c4b34fca7237", "submitted_at": "2026-08-31T03:39:43Z", "updated_at": "2026-10-03T10:41:41Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2608.30207v2" }, "canonical_source_url": "https://arxiv.org/abs/2608.30207", "catalog_id": "arxiv:2608.30207", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-31", "identifiers": { "arxiv": "2608.30207" }, "record_updated_at": "2026-10-03" }, "genre": "preprint", "id": "arxiv:2608.30207", "issued": { "date-parts": [ [ 2026, 8, 31 ] ] }, "note": "First published: 2026-08-31; Recorded arXiv version: v2; Recorded version updated: 2026-10-03T10:41:41Z.", "title": "SIR: Self-improving Red-teaming for Compute Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.00048v1", "archive": "arXiv", "archive_location": "2609.00048", "author": [ { "literal": "Lin Fu" }, { "literal": "Zheyuan Yang" }, { "literal": "Tianhui Zhang" }, { "literal": "Jinbiao Wei" }, { "literal": "Guo Gan" }, { "literal": "Boxu Liu" }, { "literal": "Yilun Zhao" }, { "literal": "Yu Rong" } ], "citation-key": "arxiv_2609_00048", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "dba8eb3f30a7e88ad973080d07fbf272f63400f4f8fa858d5f2657735ce093c0", "submitted_at": "2026-08-30T05:19:43Z", "updated_at": "2026-08-30T05:19:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.00048v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.00048", "catalog_id": "arxiv:2609.00048", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-30", "identifiers": { "arxiv": "2609.00048" }, "record_updated_at": "2026-08-30" }, "genre": "preprint", "id": "arxiv:2609.00048", "issued": { "date-parts": [ [ 2026, 8, 30 ] ] }, "note": "First published: 2026-08-30; Recorded arXiv version: v1; Recorded version updated: 2026-08-30T05:19:43Z.", "title": "GUI-CC: Benchmarking Contextual Consistency of GUI World Models as Agent Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.00524v1", "archive": "arXiv", "archive_location": "2609.00524", "author": [ { "literal": "Satwik Ram Kodandaram" }, { "literal": "Monalika Padma Reddy" }, { "literal": "Xiaojun Bi" }, { "literal": "Jiawei Zhou" }, { "literal": "I. V. Ramakrishnan" }, { "literal": "Vikas Ashok" } ], "citation-key": "arxiv_2609_00524", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "deb130d93bbbd6dfdc04c72d232a9dcfa1e92b0586831e5864070e7fe0de14dd", "submitted_at": "2026-09-01T00:44:19Z", "updated_at": "2026-09-01T00:44:19Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.00524v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.00524", "catalog_id": "arxiv:2609.00524", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-01", "identifiers": { "arxiv": "2609.00524" }, "record_updated_at": "2026-09-01" }, "genre": "preprint", "id": "arxiv:2609.00524", "issued": { "date-parts": [ [ 2026, 9, 1 ] ] }, "note": "First published: 2026-09-01; Recorded arXiv version: v1; Recorded version updated: 2026-09-01T00:44:19Z.", "title": "Are We There Yet? Assessing Computer-Use Agents for Blind Users' Accessible Interaction with Desktop Applications", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.02057v1", "archive": "arXiv", "archive_location": "2609.02057", "author": [ { "literal": "Sitong Pan" }, { "literal": "Yipeng Shen" }, { "literal": "Yilin Lu" }, { "literal": "Caiwen Ding" }, { "literal": "Lu Cheng" }, { "literal": "Qianwen Wang" } ], "citation-key": "arxiv_2609_02057", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "967f1cdda69349da3280827c0b90301e4942c44398902ab6cfb868f1d3e94a84", "submitted_at": "2026-09-02T03:34:47Z", "updated_at": "2026-09-02T03:34:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.02057v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.02057", "catalog_id": "arxiv:2609.02057", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-02", "identifiers": { "arxiv": "2609.02057" }, "record_updated_at": "2026-09-02" }, "genre": "preprint", "id": "arxiv:2609.02057", "issued": { "date-parts": [ [ 2026, 9, 2 ] ] }, "note": "First published: 2026-09-02; Recorded arXiv version: v1; Recorded version updated: 2026-09-02T03:34:47Z.", "title": "Monitoring Web Agents Without Internal Signals: Observable Trajectories and Key-Step Supervision", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.02154v1", "archive": "arXiv", "archive_location": "2609.02154", "author": [ { "literal": "Yimeng Liu" }, { "literal": "Hua Huang" } ], "citation-key": "arxiv_2609_02154", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b462dddd45e6b04a3bf6fa0bc73576a23a5f9617f61aeb257a845c0f456f731d", "submitted_at": "2026-09-02T06:06:43Z", "updated_at": "2026-09-02T06:06:43Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.02154v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.02154", "catalog_id": "arxiv:2609.02154", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-02", "identifiers": { "arxiv": "2609.02154" }, "record_updated_at": "2026-09-02" }, "genre": "preprint", "id": "arxiv:2609.02154", "issued": { "date-parts": [ [ 2026, 9, 2 ] ] }, "note": "First published: 2026-09-02; Recorded arXiv version: v1; Recorded version updated: 2026-09-02T06:06:43Z.", "title": "WiP: Characterizing and Defending Against Mobile-Agent-Driven MFA Automation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.02885v1", "archive": "arXiv", "archive_location": "2609.02885", "author": [ { "literal": "Kelvin Li" }, { "literal": "Dhruv Pendharkar" }, { "literal": "Anish Pahilajani" }, { "literal": "Chuyi Shang" }, { "literal": "Leon Oks" }, { "literal": "Leonid Karlinsky" }, { "literal": "Rogerio Feris" }, { "literal": "Trevor Darrell" }, { "literal": "Roei Herzig" } ], "citation-key": "arxiv_2609_02885", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8703474cb4f577b307731cf1a986f76f800470fde625b6ca4b3569d30e74f903", "submitted_at": "2026-09-02T17:59:40Z", "updated_at": "2026-09-02T17:59:40Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.02885v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.02885", "catalog_id": "arxiv:2609.02885", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-02", "identifiers": { "arxiv": "2609.02885" }, "record_updated_at": "2026-09-02" }, "genre": "preprint", "id": "arxiv:2609.02885", "issued": { "date-parts": [ [ 2026, 9, 2 ] ] }, "note": "First published: 2026-09-02; Recorded arXiv version: v1; Recorded version updated: 2026-09-02T17:59:40Z.", "title": "Discriminative World Models for Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.03438v1", "archive": "arXiv", "archive_location": "2609.03438", "author": [ { "literal": "Zhaoyuan Huang" }, { "literal": "Tianjie Ju" }, { "literal": "Pengzhou Cheng" }, { "literal": "Zheng Wu" }, { "literal": "Yansi Li" }, { "literal": "Chuanbiao Song" }, { "literal": "Jun Lan" }, { "literal": "Huijia Zhu" }, { "literal": "Weiqiang Wang" }, { "literal": "Zhuosheng Zhang" } ], "citation-key": "arxiv_2609_03438", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a59f7d54845de6efebf6365a188e61e41fdf31e9ebe894b1d01ce7b86d6be4ec", "submitted_at": "2026-09-03T06:48:41Z", "updated_at": "2026-09-03T06:48:41Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.03438v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.03438", "catalog_id": "arxiv:2609.03438", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-03", "identifiers": { "arxiv": "2609.03438" }, "record_updated_at": "2026-09-03" }, "genre": "preprint", "id": "arxiv:2609.03438", "issued": { "date-parts": [ [ 2026, 9, 3 ] ] }, "note": "First published: 2026-09-03; Recorded arXiv version: v1; Recorded version updated: 2026-09-03T06:48:41Z.", "title": "Do GUI Agents Know When Not to Act? Enabling Conflict-Aware Termination for Multimodal GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.04850v1", "archive": "arXiv", "archive_location": "2609.04850", "author": [ { "literal": "Weide Zhan" }, { "literal": "Qumu Shaqu" }, { "literal": "Yuanqing Liu" }, { "literal": "Peng Zhang" }, { "literal": "Jiahao Liu" }, { "literal": "Kam Him Lam" }, { "literal": "Ning Gu" }, { "literal": "Zhan Hu" }, { "literal": "Tun Lu" } ], "citation-key": "arxiv_2609_04850", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "61587798aefd8e9592ddaef756a13e02f0e4a987e4068ef582d0aaf66e8eb680", "submitted_at": "2026-09-04T08:09:47Z", "updated_at": "2026-09-04T08:09:47Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.04850v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.04850", "catalog_id": "arxiv:2609.04850", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-04", "identifiers": { "arxiv": "2609.04850" }, "record_updated_at": "2026-09-04" }, "genre": "preprint", "id": "arxiv:2609.04850", "issued": { "date-parts": [ [ 2026, 9, 4 ] ] }, "note": "First published: 2026-09-04; Recorded arXiv version: v1; Recorded version updated: 2026-09-04T08:09:47Z.", "title": "ElderBench: Benchmarking Autonomous Mobile Agents for Older Adults", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.04869v1", "archive": "arXiv", "archive_location": "2609.04869", "author": [ { "literal": "Longtao Hu" }, { "literal": "Xiao Liang" }, { "literal": "Linchao Zhu" } ], "citation-key": "arxiv_2609_04869", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "78db974038e0dde9a1347132382573fa9e2aafb1b0e1a574da3d769574329be9", "submitted_at": "2026-09-04T08:29:45Z", "updated_at": "2026-09-04T08:29:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.04869v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.04869", "catalog_id": "arxiv:2609.04869", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-04", "identifiers": { "arxiv": "2609.04869" }, "record_updated_at": "2026-09-04" }, "genre": "preprint", "id": "arxiv:2609.04869", "issued": { "date-parts": [ [ 2026, 9, 4 ] ] }, "note": "First published: 2026-09-04; Recorded arXiv version: v1; Recorded version updated: 2026-09-04T08:29:45Z.", "title": "From Interaction Traces to Persistent Skills: Online Evolution for Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.07712v1", "archive": "arXiv", "archive_location": "2609.07712", "author": [ { "literal": "Jintian Feng" }, { "literal": "Long Chen" }, { "literal": "Xiao Yu" }, { "literal": "Jiayi Dai" }, { "literal": "Chenglong Liu" }, { "literal": "Haoru Wang" }, { "literal": "Zizhen Xue" }, { "literal": "Yuxuan Shi" }, { "literal": "Ziyang Wang" }, { "literal": "Yichen Gong" } ], "citation-key": "arxiv_2609_07712", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b4090a2c7e7730ea415564b6b418d85aad77a74a222cef35435b0aa93beafd7f", "submitted_at": "2026-09-07T16:25:13Z", "updated_at": "2026-09-07T16:25:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.07712v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.07712", "catalog_id": "arxiv:2609.07712", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-07", "identifiers": { "arxiv": "2609.07712" }, "record_updated_at": "2026-09-07" }, "genre": "preprint", "id": "arxiv:2609.07712", "issued": { "date-parts": [ [ 2026, 9, 7 ] ] }, "note": "First published: 2026-09-07; Recorded arXiv version: v1; Recorded version updated: 2026-09-07T16:25:13Z.", "title": "APPSim-Bench: Bridging Real-world Apps and Reproducible Evaluation for Mobile GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.09212v1", "archive": "arXiv", "archive_location": "2609.09212", "author": [ { "literal": "Zhihao Liu" }, { "literal": "Hongyu Sun" }, { "literal": "Zhiyuan Fu" }, { "literal": "Xiaonan Duan" }, { "literal": "Jice Wang" }, { "literal": "Shangru Zhao" }, { "literal": "Weizhi Meng" }, { "literal": "Wuxin Yang" }, { "literal": "Yangfan Zhou" }, { "literal": "Yuqing Zhang" } ], "citation-key": "arxiv_2609_09212", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5a3a7a81ffe43885d9c8362df4c8aff08000304fe60477f484a732caf35124b2", "submitted_at": "2026-09-06T03:49:45Z", "updated_at": "2026-09-06T03:49:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.09212v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.09212", "catalog_id": "arxiv:2609.09212", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-06", "identifiers": { "arxiv": "2609.09212" }, "record_updated_at": "2026-09-06" }, "genre": "preprint", "id": "arxiv:2609.09212", "issued": { "date-parts": [ [ 2026, 9, 6 ] ] }, "note": "First published: 2026-09-06; Recorded arXiv version: v1; Recorded version updated: 2026-09-06T03:49:45Z.", "title": "AgentHijack: Visual Patch Attacks on Multimodal Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.10031v1", "archive": "arXiv", "archive_location": "2609.10031", "author": [ { "literal": "Xiaolei Li" }, { "literal": "Jialun Cao" }, { "literal": "Zhijian Hou" }, { "literal": "Yuzhi Zhao" }, { "literal": "Yepang Liu" }, { "literal": "Shing-Chi Cheung" } ], "citation-key": "arxiv_2609_10031", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3307d90ef09c16afaee160eacc9d2994d142be9b9a26ce9c4a57f61345e2e8ec", "submitted_at": "2026-09-09T11:04:53Z", "updated_at": "2026-09-09T11:04:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.10031v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.10031", "catalog_id": "arxiv:2609.10031", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-09", "identifiers": { "arxiv": "2609.10031" }, "record_updated_at": "2026-09-09" }, "genre": "preprint", "id": "arxiv:2609.10031", "issued": { "date-parts": [ [ 2026, 9, 9 ] ] }, "note": "First published: 2026-09-09; Recorded arXiv version: v1; Recorded version updated: 2026-09-09T11:04:53Z.", "title": "GraphDroid: Asynchronous LLM-Based Mobile App GUI Testing via History-Aware Exploration and Hybrid Intent Fulfillment", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.10451v1", "archive": "arXiv", "archive_location": "2609.10451", "author": [ { "literal": "Zixiang Chen" }, { "literal": "Yuheng Lu" }, { "literal": "Zihao Cheng" }, { "literal": "Zeming Liu" }, { "literal": "Jizeng Bai" }, { "literal": "Ziye Huang" }, { "literal": "Zhiyin Lin" }, { "literal": "Zihan Li" }, { "literal": "Yuhang Guo" }, { "literal": "Yunhong Wang" }, { "literal": "Haifeng Wang" } ], "citation-key": "arxiv_2609_10451", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5ec17516bce87a8e349983acac14415a6ceeb12c167fce57aa37d6157cd3cd83", "submitted_at": "2026-09-09T16:59:59Z", "updated_at": "2026-09-09T16:59:59Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.10451v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.10451", "catalog_id": "arxiv:2609.10451", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-09", "identifiers": { "arxiv": "2609.10451" }, "record_updated_at": "2026-09-09" }, "genre": "preprint", "id": "arxiv:2609.10451", "issued": { "date-parts": [ [ 2026, 9, 9 ] ] }, "note": "First published: 2026-09-09; Recorded arXiv version: v1; Recorded version updated: 2026-09-09T16:59:59Z.", "title": "JarvisGUI: Towards Cross-Device GUI Agents with Dynamic Task Composition", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.13873v1", "archive": "arXiv", "archive_location": "2609.13873", "author": [ { "literal": "Qihang Cen" }, { "literal": "Tianshuo Cong" }, { "literal": "Da Song" }, { "literal": "Xinlei He" }, { "literal": "Jiaxing Song" }, { "literal": "Ke Xu" }, { "literal": "Qi Li" } ], "citation-key": "arxiv_2609_13873", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "1eda33b358d4c4caf8c1fb0b1bd92698c3b37cee995d820f40ecffe82010edb4", "submitted_at": "2026-09-12T10:55:54Z", "updated_at": "2026-09-12T10:55:54Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.13873v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.13873", "catalog_id": "arxiv:2609.13873", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-12", "identifiers": { "arxiv": "2609.13873" }, "record_updated_at": "2026-09-12" }, "genre": "preprint", "id": "arxiv:2609.13873", "issued": { "date-parts": [ [ 2026, 9, 12 ] ] }, "note": "First published: 2026-09-12; Recorded arXiv version: v1; Recorded version updated: 2026-09-12T10:55:54Z.", "title": "PriMobiBench: Characterizing Visual Privacy Leakage in VLM-Driven Mobile GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.15307v1", "archive": "arXiv", "archive_location": "2609.15307", "author": [ { "literal": "Yizhou Liu" }, { "literal": "Fei Tang" }, { "literal": "Yuchen Yan" }, { "literal": "Zhengxi Lu" }, { "literal": "Songqin Nong" }, { "literal": "Tao Jiang" }, { "literal": "Wenhao Xu" }, { "literal": "Wenqi Zhang" }, { "literal": "Weiming Lu" }, { "literal": "Jun Xiao" }, { "literal": "Yongliang Shen" } ], "citation-key": "arxiv_2609_15307", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3e2081a382ae815b48eb3166d18ee7c6e53afb4cd6d75538f3ff3e29f2ed5954", "submitted_at": "2026-09-14T09:56:10Z", "updated_at": "2026-09-14T09:56:10Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.15307v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.15307", "catalog_id": "arxiv:2609.15307", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-14", "identifiers": { "arxiv": "2609.15307" }, "record_updated_at": "2026-09-14" }, "genre": "preprint", "id": "arxiv:2609.15307", "issued": { "date-parts": [ [ 2026, 9, 14 ] ] }, "note": "First published: 2026-09-14; Recorded arXiv version: v1; Recorded version updated: 2026-09-14T09:56:10Z.", "title": "Learning from Reliable Negatives: Confidence-Anchored Test-Time Adaptation for GUI Grounding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.16251v2", "archive": "arXiv", "archive_location": "2609.16251", "author": [ { "literal": "Zihan Dong" }, { "literal": "Yuanzhe Liu" }, { "literal": "Zhiyuan Ma" }, { "literal": "Qishi Zhan" }, { "literal": "Dehan Kong" }, { "literal": "Guohao Li" }, { "literal": "Kaixin Li" } ], "citation-key": "arxiv_2609_16251", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0f5fd0a3024311f9aba44a2971c9c752ee2c38444139823746b90a7c9a6d5ed4", "submitted_at": "2026-09-14T19:19:05Z", "updated_at": "2026-09-21T14:30:34Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.16251v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.16251", "catalog_id": "arxiv:2609.16251", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-14", "identifiers": { "arxiv": "2609.16251" }, "record_updated_at": "2026-09-21" }, "genre": "preprint", "id": "arxiv:2609.16251", "issued": { "date-parts": [ [ 2026, 9, 14 ] ] }, "note": "First published: 2026-09-14; Recorded arXiv version: v2; Recorded version updated: 2026-09-21T14:30:34Z.", "title": "CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.16732v1", "archive": "arXiv", "archive_location": "2609.16732", "author": [ { "literal": "Heng Li" }, { "literal": "Fulin Zhao" }, { "literal": "Zhe Geng" }, { "literal": "Zhiyuan Yao" }, { "literal": "Wei Yuan" }, { "literal": "Xiapu Luo" } ], "citation-key": "arxiv_2609_16732", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b86516b7695d23ffc9179c0829209dc28b96c3c819525eab62a293a68ab5daf0", "submitted_at": "2026-09-15T07:02:10Z", "updated_at": "2026-09-15T07:02:10Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.16732v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.16732", "catalog_id": "arxiv:2609.16732", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-15", "identifiers": { "arxiv": "2609.16732" }, "record_updated_at": "2026-09-15" }, "genre": "preprint", "id": "arxiv:2609.16732", "issued": { "date-parts": [ [ 2026, 9, 15 ] ] }, "note": "First published: 2026-09-15; Recorded arXiv version: v1; Recorded version updated: 2026-09-15T07:02:10Z.", "title": "When Agents See Differently: Exposing UI Desynchronization Threats in Mobile Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.17885v2", "archive": "arXiv", "archive_location": "2609.17885", "author": [ { "literal": "Kratika Bhagtani" }, { "literal": "Kusha Sridhar" }, { "literal": "Maziyar Baran Pouyan" }, { "literal": "Yuying Zhao" }, { "literal": "Eugene Siow" } ], "citation-key": "arxiv_2609_17885", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ab0dbea71d0b8bfe035d3b0a71d6ed99fbd261202976cdc322a0769074827d76", "submitted_at": "2026-09-15T22:23:17Z", "updated_at": "2026-09-23T01:31:01Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.17885v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.17885", "catalog_id": "arxiv:2609.17885", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-15", "identifiers": { "arxiv": "2609.17885" }, "record_updated_at": "2026-09-23" }, "genre": "preprint", "id": "arxiv:2609.17885", "issued": { "date-parts": [ [ 2026, 9, 15 ] ] }, "note": "First published: 2026-09-15; Recorded arXiv version: v2; Recorded version updated: 2026-09-23T01:31:01Z.", "title": "ERPBench: A State-Grounded Evaluation Paradigm for Computer-Use Agents in Enterprise Software", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.19523v2", "archive": "arXiv", "archive_location": "2609.19523", "author": [ { "literal": "Yinzhu Quan" }, { "literal": "Zefang Liu" } ], "citation-key": "arxiv_2609_19523", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5e7fb5704932a99147ff7c64a82a6a4041565f3edf6506a6252f75f832834d96", "submitted_at": "2026-09-17T00:29:46Z", "updated_at": "2026-10-08T04:53:11Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.19523v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.19523", "catalog_id": "arxiv:2609.19523", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-17", "identifiers": { "arxiv": "2609.19523" }, "record_updated_at": "2026-10-08" }, "genre": "preprint", "id": "arxiv:2609.19523", "issued": { "date-parts": [ [ 2026, 9, 17 ] ] }, "note": "First published: 2026-09-17; Recorded arXiv version: v2; Recorded version updated: 2026-10-08T04:53:11Z.", "title": "EconSkills: Studying Skill Transfer and Retrieval for Web Agents on Live Economic Data", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.19843v1", "archive": "arXiv", "archive_location": "2609.19843", "author": [ { "literal": "Haya Halimeh" }, { "literal": "Sascha Kaltenpoth" }, { "literal": "Kevin Bösch" }, { "literal": "Oliver Müller" } ], "citation-key": "arxiv_2609_19843", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "70a7588abcf345418ed759fb2bdb96fc60c6e1444a3e0dc98766510c755f8478", "submitted_at": "2026-09-17T07:54:15Z", "updated_at": "2026-09-17T07:54:15Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.19843v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.19843", "catalog_id": "arxiv:2609.19843", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-17", "identifiers": { "arxiv": "2609.19843" }, "record_updated_at": "2026-09-17" }, "genre": "preprint", "id": "arxiv:2609.19843", "issued": { "date-parts": [ [ 2026, 9, 17 ] ] }, "note": "First published: 2026-09-17; Recorded arXiv version: v1; Recorded version updated: 2026-09-17T07:54:15Z.", "title": "A Dual-Process Perspective on Nudge Susceptibility in LLM-Based GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.22000v2", "archive": "arXiv", "archive_location": "2609.22000", "author": [ { "literal": "Shuai Bai" }, { "literal": "Jiayong Deng" }, { "literal": "Sicheng Fan" }, { "literal": "Yikun Fu" }, { "literal": "Chang Gao" }, { "literal": "Xuhao Hu" }, { "literal": "Mianqiu Huang" }, { "literal": "Yizhen Jiang" }, { "literal": "Yuheng Jing" }, { "literal": "Dehui Kong" }, { "literal": "Keliang Li" }, { "literal": "Ning Li" }, { "literal": "Wanli Li" }, { "literal": "Dayiheng Liu" }, { "literal": "Dunjie Lu" }, { "literal": "Changwei Luo" }, { "literal": "Que Shen" }, { "literal": "Zheyuan Wang" }, { "literal": "Zijian Wang" }, { "literal": "Jie Wu" }, { "literal": "Gao Wu" }, { "literal": "Zhihui Xie" }, { "literal": "Rui Xie" }, { "literal": "Haiyang Xu" }, { "literal": "An Yang" }, { "literal": "Jiakang Yuan" }, { "literal": "Yanming Zhang" }, { "literal": "Jiajun Zhang" }, { "literal": "Xi Zhang" }, { "literal": "Zhenru Zhang" }, { "literal": "Zhuo Zhen" }, { "literal": "Mingkang Zhu" }, { "literal": "Bowen Zhou" } ], "citation-key": "arxiv_2609_22000", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3b312c5a4d638ec9f9a7408fd9ce2d8cbf74fa462d48c8aa37f59e4ad3bd8edc", "submitted_at": "2026-09-18T17:00:56Z", "updated_at": "2026-09-21T07:09:10Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.22000v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.22000", "catalog_id": "arxiv:2609.22000", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-18", "identifiers": { "arxiv": "2609.22000" }, "record_updated_at": "2026-09-21" }, "genre": "preprint", "id": "arxiv:2609.22000", "issued": { "date-parts": [ [ 2026, 9, 18 ] ] }, "note": "First published: 2026-09-18; Recorded arXiv version: v2; Recorded version updated: 2026-09-21T07:09:10Z.", "title": "RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.24277v1", "archive": "arXiv", "archive_location": "2609.24277", "author": [ { "literal": "Yunxiang Li" }, { "literal": "Xixin Wu" }, { "literal": "Helen Meng" } ], "citation-key": "arxiv_2609_24277", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "64936af65d0ca351b19da3f63325f2bc2a5c9e0951e5a7806cd4dd82b18784a7", "submitted_at": "2026-09-21T08:42:08Z", "updated_at": "2026-09-21T08:42:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.24277v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.24277", "catalog_id": "arxiv:2609.24277", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-21", "identifiers": { "arxiv": "2609.24277" }, "record_updated_at": "2026-09-21" }, "genre": "preprint", "id": "arxiv:2609.24277", "issued": { "date-parts": [ [ 2026, 9, 21 ] ] }, "note": "First published: 2026-09-21; Recorded arXiv version: v1; Recorded version updated: 2026-09-21T08:42:08Z.", "title": "How Many Pixels Is a Digit Worth? Place-Aware Coordinate Entropy for GUI Agent Confidence Estimation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.24890v1", "archive": "arXiv", "archive_location": "2609.24890", "author": [ { "literal": "Zhilin Wang" }, { "literal": "Shaokun Zhang" }, { "literal": "Yifan Zhang" }, { "literal": "Hao Zhang" }, { "literal": "Jin Xu" }, { "literal": "Binfeng Xu" }, { "literal": "Jian Hu" }, { "literal": "Yunheng Zou" }, { "literal": "Karan Sapra" }, { "literal": "Andrew Tao" }, { "literal": "Jan Kautz" }, { "literal": "Yi Dong" } ], "citation-key": "arxiv_2609_24890", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0d3cb920a677108910530788eb68c263127dba9510274f182910aed9c2f848c9", "submitted_at": "2026-09-21T16:55:24Z", "updated_at": "2026-09-21T16:55:24Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.24890v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.24890", "catalog_id": "arxiv:2609.24890", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-21", "identifiers": { "arxiv": "2609.24890" }, "record_updated_at": "2026-09-21" }, "genre": "preprint", "id": "arxiv:2609.24890", "issued": { "date-parts": [ [ 2026, 9, 21 ] ] }, "note": "First published: 2026-09-21; Recorded arXiv version: v1; Recorded version updated: 2026-09-21T16:55:24Z.", "title": "OSWorld-Pro: Process-based Evaluation for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.27273v2", "archive": "arXiv", "archive_location": "2609.27273", "author": [ { "literal": "Yuxuan Li" }, { "literal": "Will Epperson" }, { "literal": "Wesley Deng" }, { "literal": "Zezhou Huang" } ], "citation-key": "arxiv_2609_27273", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8d241c1ea9731bc815eef5df9c409d410608adb052616406d955d7d33cecd9b9", "submitted_at": "2026-09-23T03:00:00Z", "updated_at": "2026-09-27T20:49:01Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.27273v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.27273", "catalog_id": "arxiv:2609.27273", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-23", "identifiers": { "arxiv": "2609.27273" }, "record_updated_at": "2026-09-27" }, "genre": "preprint", "id": "arxiv:2609.27273", "issued": { "date-parts": [ [ 2026, 9, 23 ] ] }, "note": "First published: 2026-09-23; Recorded arXiv version: v2; Recorded version updated: 2026-09-27T20:49:01Z.", "title": "CAVEAT: Towards Robust Computer-Use Agents in Incentive-Misaligned Environments", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.27353v1", "archive": "arXiv", "archive_location": "2609.27353", "author": [ { "literal": "Chengguang Gan" }, { "literal": "Yunhao Liang" }, { "literal": "QingHao Zhang" }, { "literal": "Shiwen Ni" } ], "citation-key": "arxiv_2609_27353", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2b2bda1c50fd47fabbe77c40a2938979a6a16b407b40f11737fc6f1fc795d090", "submitted_at": "2026-09-23T04:49:07Z", "updated_at": "2026-09-23T04:49:07Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.27353v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.27353", "catalog_id": "arxiv:2609.27353", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-23", "identifiers": { "arxiv": "2609.27353" }, "record_updated_at": "2026-09-23" }, "genre": "preprint", "id": "arxiv:2609.27353", "issued": { "date-parts": [ [ 2026, 9, 23 ] ] }, "note": "First published: 2026-09-23; Recorded arXiv version: v1; Recorded version updated: 2026-09-23T04:49:07Z.", "title": "Guides That Cause Actions: An Offline Study of Guide-Action Mutual Reinforcement in Multimodal Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.27461v1", "archive": "arXiv", "archive_location": "2609.27461", "author": [ { "literal": "Zeyu Zhang" }, { "literal": "Dingyi Rong" }, { "literal": "Zijian Chen" }, { "literal": "Zicheng Zhang" }, { "literal": "Xiongkuo Min" }, { "literal": "Guangtao Zhai" } ], "citation-key": "arxiv_2609_27461", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3a89489bb1ff3001a1d806c3eefd8a6aa19503d1dfbe0265b916766119f96ec2", "submitted_at": "2026-09-23T07:27:23Z", "updated_at": "2026-09-23T07:27:23Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.27461v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.27461", "catalog_id": "arxiv:2609.27461", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-23", "identifiers": { "arxiv": "2609.27461" }, "record_updated_at": "2026-09-23" }, "genre": "preprint", "id": "arxiv:2609.27461", "issued": { "date-parts": [ [ 2026, 9, 23 ] ] }, "note": "First published: 2026-09-23; Recorded arXiv version: v1; Recorded version updated: 2026-09-23T07:27:23Z.", "title": "Invisible in Space, Visible in Time: Motion Vision CAPTCHA against GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.27770v1", "archive": "arXiv", "archive_location": "2609.27770", "author": [ { "literal": "Zhitong Guo" }, { "literal": "Jing Yu Koh" }, { "literal": "Ruiyu Li" } ], "citation-key": "arxiv_2609_27770", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "80f6fb973e9f8d62e515d02105d08be774fc32ed5b20a7ecee92b45d112415aa", "submitted_at": "2026-08-16T19:54:53Z", "updated_at": "2026-08-16T19:54:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.27770v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.27770", "catalog_id": "arxiv:2609.27770", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-16", "identifiers": { "arxiv": "2609.27770" }, "record_updated_at": "2026-08-16" }, "genre": "preprint", "id": "arxiv:2609.27770", "issued": { "date-parts": [ [ 2026, 8, 16 ] ] }, "note": "First published: 2026-08-16; Recorded arXiv version: v1; Recorded version updated: 2026-08-16T19:54:53Z.", "title": "Improving LLM-based Autonomous Web Agents with Filtering", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.30604v1", "archive": "arXiv", "archive_location": "2609.30604", "author": [ { "literal": "Alexander Gill" }, { "literal": "Md Farhan Ishmam" }, { "literal": "Xuyen Nguyen" }, { "literal": "Neha Bhat" }, { "literal": "Parker Henry DeYoung" }, { "literal": "Fateme Hashemi Chaleshtori" }, { "literal": "Nathan Stringham" }, { "literal": "Kenneth Marino" }, { "literal": "Ana Marasović" } ], "citation-key": "arxiv_2609_30604", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d6ed206d8f8fec53c03956fffcafec6b55d94a47fc705bbc8b77c832b76419e6", "submitted_at": "2026-09-24T22:36:10Z", "updated_at": "2026-09-24T22:36:10Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.30604v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.30604", "catalog_id": "arxiv:2609.30604", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-24", "identifiers": { "arxiv": "2609.30604" }, "record_updated_at": "2026-09-24" }, "genre": "preprint", "id": "arxiv:2609.30604", "issued": { "date-parts": [ [ 2026, 9, 24 ] ] }, "note": "First published: 2026-09-24; Recorded arXiv version: v1; Recorded version updated: 2026-09-24T22:36:10Z.", "title": "The Hard Part Comes After Search: Benchmarking Web Agents on Synthesizing, Organizing, and Displaying Knowledge", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.31364v1", "archive": "arXiv", "archive_location": "2609.31364", "author": [ { "literal": "Kiana Hooshanfar" }, { "literal": "Amirhossein Kazerouni" }, { "literal": "Alireza Hosseini" }, { "literal": "Michael Brudno" }, { "literal": "Babak Taati" } ], "citation-key": "arxiv_2609_31364", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9ce76f34abe391289643321027a33ac3cac8b5bd64eaa901f9a5fbf7c5561362", "submitted_at": "2026-09-25T14:59:14Z", "updated_at": "2026-09-25T14:59:14Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.31364v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.31364", "catalog_id": "arxiv:2609.31364", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-09-25", "identifiers": { "arxiv": "2609.31364" }, "record_updated_at": "2026-09-25" }, "genre": "preprint", "id": "arxiv:2609.31364", "issued": { "date-parts": [ [ 2026, 9, 25 ] ] }, "note": "First published: 2026-09-25; Recorded arXiv version: v1; Recorded version updated: 2026-09-25T14:59:14Z.", "title": "OpenVAM: Open-World Visual Attention Modeling with VLMs", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.32036v1", "archive": "arXiv", "archive_location": "2609.32036", "author": [ { "literal": "Chenyue Li" }, { "literal": "Xiaoxiao Sun" }, { "literal": "Yubo Deng" }, { "literal": "Qinlin Zhao" }, { "literal": "Serena Yeung-Levy" }, { "literal": "Yuhui Zhang" } ], "citation-key": "arxiv_2609_32036", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "12976e58eb48d9c4709fb03508ac7ed7eeafffd3df675d3dccfd09bdc354f3db", "submitted_at": "2026-09-25T22:07:29Z", "updated_at": "2026-09-25T22:07:29Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.32036v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.32036", "catalog_id": "arxiv:2609.32036", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-25", "identifiers": { "arxiv": "2609.32036" }, "record_updated_at": "2026-09-25" }, "genre": "preprint", "id": "arxiv:2609.32036", "issued": { "date-parts": [ [ 2026, 9, 25 ] ] }, "note": "First published: 2026-09-25; Recorded arXiv version: v1; Recorded version updated: 2026-09-25T22:07:29Z.", "title": "ScreenHaystack: Finding Blind Zones in GUI Grounding", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.32385v1", "archive": "arXiv", "archive_location": "2609.32385", "author": [ { "literal": "Chuhan Zhang" }, { "literal": "Qi Xie" }, { "literal": "Ziyue Wang" }, { "literal": "Jianing Yin" }, { "literal": "Yunfan Zhou" }, { "literal": "Dazhen Deng" }, { "literal": "Yingcai Wu" } ], "citation-key": "arxiv_2609_32385", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f091cab38c932e08d21afa936dfe241c93c0b2139aba1d1ce7a6d1513ad5da1a", "submitted_at": "2026-09-26T08:58:50Z", "updated_at": "2026-09-26T08:58:50Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.32385v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.32385", "catalog_id": "arxiv:2609.32385", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-26", "identifiers": { "arxiv": "2609.32385" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2609.32385", "issued": { "date-parts": [ [ 2026, 9, 26 ] ] }, "note": "First published: 2026-09-26; Recorded arXiv version: v1; Recorded version updated: 2026-09-26T08:58:50Z.", "title": "DashAct: A Progressive Diagnostic Benchmark for GUI Agents in Interactive Dashboard Analysis", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.32511v1", "archive": "arXiv", "archive_location": "2609.32511", "author": [ { "literal": "Jinming Hu" }, { "literal": "Haodong Zhao" }, { "literal": "Qi Jia" }, { "literal": "Die Chen" }, { "literal": "Tianhang Zhao" }, { "literal": "Sufeng Duan" }, { "literal": "Gongshen Liu" } ], "citation-key": "arxiv_2609_32511", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ff1c4eff95094ea8ecf37b7bc621eebbd5e4a677b0e11240e802a72b9b229c7f", "submitted_at": "2026-09-26T11:59:39Z", "updated_at": "2026-09-26T11:59:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.32511v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.32511", "catalog_id": "arxiv:2609.32511", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-26", "identifiers": { "arxiv": "2609.32511" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2609.32511", "issued": { "date-parts": [ [ 2026, 9, 26 ] ] }, "note": "First published: 2026-09-26; Recorded arXiv version: v1; Recorded version updated: 2026-09-26T11:59:39Z.", "title": "Learning from Others, Acting for You: Cross-User Memory Sharing for LLM Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.32600v1", "archive": "arXiv", "archive_location": "2609.32600", "author": [ { "literal": "Prince Zizhuang Wang" }, { "literal": "Chenhao Liang" }, { "literal": "Zelong Xu" }, { "literal": "Aojie Yuan" }, { "literal": "Xiaolin Zhou" }, { "literal": "Haiyue Zhang" }, { "literal": "Yue Zhao" }, { "literal": "Xiyang Hu" }, { "literal": "Shuli Jiang" } ], "citation-key": "arxiv_2609_32600", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d53f2512b99828849f047f63412264437a59b730c117c801de252e6dc80a33fc", "submitted_at": "2026-09-26T13:18:00Z", "updated_at": "2026-09-26T13:18:00Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.32600v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.32600", "catalog_id": "arxiv:2609.32600", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-26", "identifiers": { "arxiv": "2609.32600" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2609.32600", "issued": { "date-parts": [ [ 2026, 9, 26 ] ] }, "note": "First published: 2026-09-26; Recorded arXiv version: v1; Recorded version updated: 2026-09-26T13:18:00Z.", "title": "CUA-SWE: When Computer-Use Agents Meet Visual Software Engineering", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.32679v1", "archive": "arXiv", "archive_location": "2609.32679", "author": [ { "literal": "Dongsheng Liu" }, { "literal": "Chao Jin" }, { "literal": "Wenkui Yang" }, { "literal": "Hejin Wang" }, { "literal": "Junwei Yang" }, { "literal": "Zeren Zhang" }, { "literal": "Ziwei Chen" }, { "literal": "Huaibo Huang" }, { "literal": "Jie Cao" }, { "literal": "Ran He" } ], "citation-key": "arxiv_2609_32679", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d641819d168eedc07270607da33f3197c15f2f90cf99f538c3839eed506f3308", "submitted_at": "2026-09-26T14:35:45Z", "updated_at": "2026-09-26T14:35:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.32679v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.32679", "catalog_id": "arxiv:2609.32679", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-26", "identifiers": { "arxiv": "2609.32679" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2609.32679", "issued": { "date-parts": [ [ 2026, 9, 26 ] ] }, "note": "First published: 2026-09-26; Recorded arXiv version: v1; Recorded version updated: 2026-09-26T14:35:45Z.", "title": "The GUI Is Not the State: Diagnosing State Aliasing in GUI World Models", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.32915v1", "archive": "arXiv", "archive_location": "2609.32915", "author": [ { "literal": "Asif Shahriar" }, { "literal": "Md Nafiu Rahman" }, { "literal": "Sadif Ahmed" }, { "literal": "Farig Sadeque" }, { "literal": "Md Rizwan Parvez" } ], "citation-key": "arxiv_2609_32915", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "802342ce2c486c30c84c33c897becb0348831034d4d11f8d3d3861ec0380b65c", "submitted_at": "2026-09-26T20:09:16Z", "updated_at": "2026-09-26T20:09:16Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.32915v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.32915", "catalog_id": "arxiv:2609.32915", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-26", "identifiers": { "arxiv": "2609.32915" }, "record_updated_at": "2026-09-26" }, "genre": "preprint", "id": "arxiv:2609.32915", "issued": { "date-parts": [ [ 2026, 9, 26 ] ] }, "note": "First published: 2026-09-26; Recorded arXiv version: v1; Recorded version updated: 2026-09-26T20:09:16Z.", "title": "AgentTell: Behavioural Side-Channel Leakage in Browser-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.33102v1", "archive": "arXiv", "archive_location": "2609.33102", "author": [ { "literal": "Ben Hagag" }, { "literal": "William L. Anderson" }, { "literal": "Srija Chakraborty" }, { "literal": "Christian Schroeder de Witt" } ], "citation-key": "arxiv_2609_33102", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3c380079220050fbe354509238a5a79c7df2917a245490e4ec0cff512f7565ab", "submitted_at": "2026-09-27T02:20:08Z", "updated_at": "2026-09-27T02:20:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.33102v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.33102", "catalog_id": "arxiv:2609.33102", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-27", "identifiers": { "arxiv": "2609.33102" }, "record_updated_at": "2026-09-27" }, "genre": "preprint", "id": "arxiv:2609.33102", "issued": { "date-parts": [ [ 2026, 9, 27 ] ] }, "note": "First published: 2026-09-27; Recorded arXiv version: v1; Recorded version updated: 2026-09-27T02:20:08Z.", "title": "ORBIT: A Framework for Multi-Agent Safety and Security Evaluations", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.34113v2", "archive": "arXiv", "archive_location": "2609.34113", "author": [ { "literal": "Shaoqing Zhang" }, { "literal": "Kehai Chen" }, { "literal": "Xuefeng Bai" }, { "literal": "Zhuosheng Zhang" }, { "literal": "Pengfei Zhang" }, { "literal": "Yang Xiang" }, { "literal": "Min Zhang" } ], "citation-key": "arxiv_2609_34113", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2f4116455785579ece9a2ff1bae304b62f1cbe13a91db8d146bd281a27d19eb0", "submitted_at": "2026-09-28T01:48:48Z", "updated_at": "2026-09-29T09:05:42Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.34113v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.34113", "catalog_id": "arxiv:2609.34113", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-28", "identifiers": { "arxiv": "2609.34113" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2609.34113", "issued": { "date-parts": [ [ 2026, 9, 28 ] ] }, "note": "First published: 2026-09-28; Recorded arXiv version: v2; Recorded version updated: 2026-09-29T09:05:42Z.", "title": "GUITAR: Structured Failure Diagnosis of GUI Agents via State Transitions", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.34139v1", "archive": "arXiv", "archive_location": "2609.34139", "author": [ { "literal": "Tien Tran" }, { "literal": "Namho Koh" }, { "literal": "Daiki E. Matsunaga" }, { "literal": "Ayush Jain" }, { "literal": "Kee Eung Kim" } ], "citation-key": "arxiv_2609_34139", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0ee7ddc170afcd2fa09d1c8d12a07bbd6aac692bc12ffc27bd39f1de0c912dfa", "submitted_at": "2026-09-28T02:21:13Z", "updated_at": "2026-09-28T02:21:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.34139v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.34139", "catalog_id": "arxiv:2609.34139", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-28", "identifiers": { "arxiv": "2609.34139" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2609.34139", "issued": { "date-parts": [ [ 2026, 9, 28 ] ] }, "note": "First published: 2026-09-28; Recorded arXiv version: v1; Recorded version updated: 2026-09-28T02:21:13Z.", "title": "Same Tasks, Different Apps: Why Mobile GUI Agents Fail to Generalize?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.34321v1", "archive": "arXiv", "archive_location": "2609.34321", "author": [ { "literal": "Ziqiang Wang" }, { "literal": "Li Gu" }, { "literal": "Zhixiang Chi" }, { "literal": "Linlian Jiang" }, { "literal": "Zihuan Jiang" }, { "literal": "Linqiang Guo" }, { "literal": "Siobhan Reid" }, { "literal": "Zhi Liu" }, { "literal": "Yang Wang" } ], "citation-key": "arxiv_2609_34321", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7fbaeb0b4f891d64cc12a3d5663b6804df6c7db82e6a7ccfd54aa10e837010de", "submitted_at": "2026-09-28T05:00:45Z", "updated_at": "2026-09-28T05:00:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.34321v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.34321", "catalog_id": "arxiv:2609.34321", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-28", "identifiers": { "arxiv": "2609.34321" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2609.34321", "issued": { "date-parts": [ [ 2026, 9, 28 ] ] }, "note": "First published: 2026-09-28; Recorded arXiv version: v1; Recorded version updated: 2026-09-28T05:00:45Z.", "title": "One Rollout Is All You Get: Fully Test-Time Adaptation for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.34769v1", "archive": "arXiv", "archive_location": "2609.34769", "author": [ { "literal": "Bingo Zhang" }, { "literal": "Haochuan Lu" }, { "literal": "Zongjie Li" }, { "literal": "Genjian Li" }, { "literal": "Ari Yu Zhang" }, { "literal": "Chaozheng Wang" } ], "citation-key": "arxiv_2609_34769", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f81fd07262dccf6783729e4336fbd77835f2fcdea030eef8dc2272e7ef82b8fc", "submitted_at": "2026-09-28T09:49:25Z", "updated_at": "2026-09-28T09:49:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.34769v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.34769", "catalog_id": "arxiv:2609.34769", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-09-28", "identifiers": { "arxiv": "2609.34769" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2609.34769", "issued": { "date-parts": [ [ 2026, 9, 28 ] ] }, "note": "First published: 2026-09-28; Recorded arXiv version: v1; Recorded version updated: 2026-09-28T09:49:25Z.", "title": "LongPuzzleBench: Evaluating GUI Agents on Long-Horizon Visual Puzzles", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.35026v1", "archive": "arXiv", "archive_location": "2609.35026", "author": [ { "literal": "Anton Emelyanov" }, { "literal": "Maria Tikhonova" }, { "literal": "Zaven Martirosian" }, { "literal": "Sergei Averkiev" }, { "literal": "Alena Fenogenova" } ], "citation-key": "arxiv_2609_35026", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d5090c6b7879e38102b1180ee1e4dfd1552af777172e2f94bc0f9a684fb59dc7", "submitted_at": "2026-09-28T12:24:23Z", "updated_at": "2026-09-28T12:24:23Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.35026v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.35026", "catalog_id": "arxiv:2609.35026", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-28", "identifiers": { "arxiv": "2609.35026" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2609.35026", "issued": { "date-parts": [ [ 2026, 9, 28 ] ] }, "note": "First published: 2026-09-28; Recorded arXiv version: v1; Recorded version updated: 2026-09-28T12:24:23Z.", "title": "WebPageBench: Event-Level Verification and Controlled UI-Variant Generation for Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.35814v1", "archive": "arXiv", "archive_location": "2609.35814", "author": [ { "literal": "Xunjian Yin" }, { "literal": "Tianchen Guan" }, { "literal": "Jinao Wang" }, { "literal": "Weili Cao" }, { "literal": "Daisy Xinlei Lin" }, { "literal": "Royce Cheng-Yue" }, { "literal": "Keagan Long" }, { "literal": "Kyle Wong" }, { "literal": "Bhuwan Dhingra" }, { "literal": "Xiangjun Wang" }, { "literal": "Shuyan Zhou" } ], "citation-key": "arxiv_2609_35814", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "197cfb8424584955bc85602d6559037ad31ee815f89b45dc8f9f18c8636552ad", "submitted_at": "2026-09-20T16:16:23Z", "updated_at": "2026-09-20T16:16:23Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.35814v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.35814", "catalog_id": "arxiv:2609.35814", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-20", "identifiers": { "arxiv": "2609.35814" }, "record_updated_at": "2026-09-20" }, "genre": "preprint", "id": "arxiv:2609.35814", "issued": { "date-parts": [ [ 2026, 9, 20 ] ] }, "note": "First published: 2026-09-20; Recorded arXiv version: v1; Recorded version updated: 2026-09-20T16:16:23Z.", "title": "Constructing Challenging Browser-Use Tasks by Controlled Environment Interventions", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.36201v1", "archive": "arXiv", "archive_location": "2609.36201", "author": [ { "literal": "Jianxing Chen" }, { "literal": "Xiao Yu" }, { "literal": "Shipra Agrawal" }, { "literal": "Zhou Yu" } ], "citation-key": "arxiv_2609_36201", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e92c63d4cf58375aa7068570f8a98c145e24f8d393c5807c0aff3ed67f18b9f0", "submitted_at": "2026-09-28T20:01:35Z", "updated_at": "2026-09-28T20:01:35Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.36201v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.36201", "catalog_id": "arxiv:2609.36201", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-28", "identifiers": { "arxiv": "2609.36201" }, "record_updated_at": "2026-09-28" }, "genre": "preprint", "id": "arxiv:2609.36201", "issued": { "date-parts": [ [ 2026, 9, 28 ] ] }, "note": "First published: 2026-09-28; Recorded arXiv version: v1; Recorded version updated: 2026-09-28T20:01:35Z.", "title": "SCOUT: Synergizing Reasoning and Tool-Use for Computer-Use Safety", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.36488v1", "archive": "arXiv", "archive_location": "2609.36488", "author": [ { "literal": "Dongchan Shin" }, { "literal": "Xing Han Lù" }, { "literal": "Jiaqi Deng" }, { "literal": "Jay Gala" }, { "literal": "Tomás Vergara Browne" }, { "literal": "Jaewon Moon" }, { "literal": "Fengyuan Liu" }, { "literal": "Alexandre Drouin" }, { "literal": "Siva Reddy" }, { "literal": "Alexandre Lacoste" } ], "citation-key": "arxiv_2609_36488", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2fe39365133228daa770445f19002f6abe34ca1151cf4952a225ba0001de5250", "submitted_at": "2026-09-29T01:46:48Z", "updated_at": "2026-09-29T01:46:48Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.36488v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.36488", "catalog_id": "arxiv:2609.36488", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-29", "identifiers": { "arxiv": "2609.36488" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2609.36488", "issued": { "date-parts": [ [ 2026, 9, 29 ] ] }, "note": "First published: 2026-09-29; Recorded arXiv version: v1; Recorded version updated: 2026-09-29T01:46:48Z.", "title": "AdaptArena: Evaluating Test-Time Personalization of Web Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.36939v1", "archive": "arXiv", "archive_location": "2609.36939", "author": [ { "literal": "Shengtian Yang" }, { "literal": "Ziyu Xiong" }, { "literal": "Kaibing Yang" }, { "literal": "Guangfeng Cai" }, { "literal": "Yewen Li" }, { "literal": "Peng Jiang" }, { "literal": "Gai Kun" }, { "literal": "Qingpeng Cai" }, { "literal": "Lei Feng" } ], "citation-key": "arxiv_2609_36939", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "d45cee8117e302aa190d38c8298bf479936a5a27aa9b85716a76a2a90499069c", "submitted_at": "2026-09-29T07:51:57Z", "updated_at": "2026-09-29T07:51:57Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.36939v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.36939", "catalog_id": "arxiv:2609.36939", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-29", "identifiers": { "arxiv": "2609.36939" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2609.36939", "issued": { "date-parts": [ [ 2026, 9, 29 ] ] }, "note": "First published: 2026-09-29; Recorded arXiv version: v1; Recorded version updated: 2026-09-29T07:51:57Z.", "title": "SCA: Spatial Credit Assignment for Reinforcement Learning of GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.37053v1", "archive": "arXiv", "archive_location": "2609.37053", "author": [ { "literal": "Mei Wu" }, { "literal": "Rui Xie" }, { "literal": "Runyu Zhang" }, { "literal": "Yuqiang Li" }, { "literal": "Tianfan Fu" }, { "literal": "Bo Chen" }, { "literal": "Kai Yu" }, { "literal": "Xin Chen" }, { "literal": "Lu Chen" } ], "citation-key": "arxiv_2609_37053", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e032f67257a52dbc59310d921d802e4dc85af3af110d22bb6d005d202bbaa6b2", "submitted_at": "2026-09-29T08:58:51Z", "updated_at": "2026-09-29T08:58:51Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.37053v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.37053", "catalog_id": "arxiv:2609.37053", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-29", "identifiers": { "arxiv": "2609.37053" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2609.37053", "issued": { "date-parts": [ [ 2026, 9, 29 ] ] }, "note": "First published: 2026-09-29; Recorded arXiv version: v1; Recorded version updated: 2026-09-29T08:58:51Z.", "title": "MatToolBench: Benchmarking Multimodal Agents in Real-World Materials Science Workflows", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.37686v2", "archive": "arXiv", "archive_location": "2609.37686", "author": [ { "literal": "Hongcheng Gao" }, { "literal": "Hailong Qu" }, { "literal": "Yu Lei" }, { "literal": "Henghui Sun" }, { "literal": "Haoyang Li" }, { "literal": "Yipeng Wei" }, { "literal": "Naihao Xue" }, { "literal": "Xiaohan Yu" }, { "literal": "Zhuo Tao" }, { "literal": "Yihe Zang" }, { "literal": "Yajiao Wang" }, { "literal": "Jingyi Tang" }, { "literal": "Yi Li" }, { "literal": "Jingjing Zhou" }, { "literal": "Jie Luo" }, { "literal": "Bohan Zeng" }, { "literal": "Chengyu Shen" }, { "literal": "Hao Jiang" }, { "literal": "Chong Chen" }, { "literal": "Bowen Qu" }, { "literal": "Olive Huang" }, { "literal": "Zeqiang Wang" } ], "citation-key": "arxiv_2609_37686", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "aab7f586cb58a1f2d20267fd810ba34d6d4d166e3da715d2b3ff0b77e464473e", "submitted_at": "2026-09-29T14:32:35Z", "updated_at": "2026-10-09T08:15:21Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.37686v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.37686", "catalog_id": "arxiv:2609.37686", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-29", "identifiers": { "arxiv": "2609.37686" }, "record_updated_at": "2026-10-09" }, "genre": "preprint", "id": "arxiv:2609.37686", "issued": { "date-parts": [ [ 2026, 9, 29 ] ] }, "note": "First published: 2026-09-29; Recorded arXiv version: v2; Recorded version updated: 2026-10-09T08:15:21Z.", "title": "EngiWorld: What Can Frontier Agents Deliver in Professional Engineering Environments?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.38184v1", "archive": "arXiv", "archive_location": "2609.38184", "author": [ { "literal": "Yongjiang Wu" }, { "literal": "Junyuan Zhang" }, { "literal": "Ada Chen" }, { "literal": "Kuiyi Gao" }, { "literal": "Wenxuan Wang" } ], "citation-key": "arxiv_2609_38184", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f14844abefab6b4eb464a1c7675f9de7f460849872ebb776230724ed5828feed", "submitted_at": "2026-08-07T07:51:13Z", "updated_at": "2026-08-07T07:51:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.38184v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.38184", "catalog_id": "arxiv:2609.38184", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-08-07", "identifiers": { "arxiv": "2609.38184" }, "record_updated_at": "2026-08-07" }, "genre": "preprint", "id": "arxiv:2609.38184", "issued": { "date-parts": [ [ 2026, 8, 7 ] ] }, "note": "First published: 2026-08-07; Recorded arXiv version: v1; Recorded version updated: 2026-08-07T07:51:13Z.", "title": "RealGUINoise: An Interactive Cross-Platform Benchmark for GUI Agent Robustness under Real-World Interface Noise", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.38867v1", "archive": "arXiv", "archive_location": "2609.38867", "author": [ { "literal": "Terumi Chiba" }, { "literal": "Guangzhi Sun" }, { "literal": "Zheqi Yuan" }, { "literal": "Chao Zhang" } ], "citation-key": "arxiv_2609_38867", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "bded059a37101b0e6a91ba662e0d70e5e010ddb8c3557598dff2ba82afa45e34", "submitted_at": "2026-09-30T03:19:25Z", "updated_at": "2026-09-30T03:19:25Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.38867v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.38867", "catalog_id": "arxiv:2609.38867", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-30", "identifiers": { "arxiv": "2609.38867" }, "record_updated_at": "2026-09-30" }, "genre": "preprint", "id": "arxiv:2609.38867", "issued": { "date-parts": [ [ 2026, 9, 30 ] ] }, "note": "First published: 2026-09-30; Recorded arXiv version: v1; Recorded version updated: 2026-09-30T03:19:25Z.", "title": "Talk2Agent: Benchmarking Voice Interfaces for Text Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.39903v1", "archive": "arXiv", "archive_location": "2609.39903", "author": [ { "literal": "Dingyuan Dai" }, { "literal": "Heli Qi" }, { "literal": "Lei Liu" }, { "literal": "Yinxi Li" }, { "literal": "Baiding Chen" }, { "literal": "Zijun Dou" }, { "literal": "Qingcheng Zeng" }, { "literal": "Qi Kang" }, { "literal": "Oliver Sun" }, { "literal": "Eric Wang" }, { "literal": "Bo Zhou" }, { "literal": "Haixin Wang" }, { "literal": "Yufan Du" }, { "literal": "Shi Bo" }, { "literal": "Ruihan Lin" }, { "literal": "Mengqi Yuan" }, { "literal": "Dunjie Lu" }, { "literal": "Steven Dillmann" }, { "literal": "Yiming Shi" }, { "literal": "Tina Su" }, { "literal": "Amy Xin" }, { "literal": "Minghao Liu" }, { "literal": "Xi Wang" }, { "literal": "Xu Huang" }, { "literal": "Ge Zhang" }, { "literal": "Pengyu Nie" }, { "literal": "Zhen Yang" }, { "literal": "Jie Tang" }, { "literal": "Juanzi Li" }, { "literal": "Weihao Xuan" }, { "literal": "Tianyu Liu" } ], "citation-key": "arxiv_2609_39903", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "690cd75934bae65311240cd4e6db7b8fd607056bd60799bf70cddd03efda9261", "submitted_at": "2026-09-30T15:03:05Z", "updated_at": "2026-09-30T15:03:05Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.39903v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.39903", "catalog_id": "arxiv:2609.39903", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-30", "identifiers": { "arxiv": "2609.39903" }, "record_updated_at": "2026-09-30" }, "genre": "preprint", "id": "arxiv:2609.39903", "issued": { "date-parts": [ [ 2026, 9, 30 ] ] }, "note": "First published: 2026-09-30; Recorded arXiv version: v1; Recorded version updated: 2026-09-30T15:03:05Z.", "title": "OSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific Software", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2609.40253v2", "archive": "arXiv", "archive_location": "2609.40253", "author": [ { "literal": "Yong Du" }, { "literal": "Tongbo Chen" }, { "literal": "Zhengxi Lu" }, { "literal": "Yizhou Liu" }, { "literal": "Bofan Chen" }, { "literal": "Tao Jiang" }, { "literal": "Wenhao Xu" }, { "literal": "Yongliang Shen" } ], "citation-key": "arxiv_2609_40253", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "8f3745b6b497c8819dc990bdb5d74e7af463d23654cbc7cdd76ade0faf2b3e0c", "submitted_at": "2026-09-30T17:40:09Z", "updated_at": "2026-10-01T17:50:14Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2609.40253v2" }, "canonical_source_url": "https://arxiv.org/abs/2609.40253", "catalog_id": "arxiv:2609.40253", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-30", "identifiers": { "arxiv": "2609.40253" }, "record_updated_at": "2026-10-01" }, "genre": "preprint", "id": "arxiv:2609.40253", "issued": { "date-parts": [ [ 2026, 9, 30 ] ] }, "note": "First published: 2026-09-30; Recorded arXiv version: v2; Recorded version updated: 2026-10-01T17:50:14Z.", "title": "ComputerSD: Online Self-Distillation from Real-Time Feedback for Computer-Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2609.40284v1", "archive": "arXiv", "archive_location": "2609.40284", "author": [ { "literal": "Pranjal Aggarwal" }, { "literal": "Lawrence Keunho Jang" }, { "literal": "Sean Welleck" }, { "literal": "Daniel Fried" }, { "literal": "Ruslan Salakhutdinov" }, { "literal": "Jing Yu Koh" } ], "citation-key": "arxiv_2609_40284", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "502912cdc35afd61beeb0202d55fc8feefcd92cd3719cd0e85bc71561d6f5a3c", "submitted_at": "2026-09-30T17:48:06Z", "updated_at": "2026-09-30T17:48:06Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2609.40284v1" }, "canonical_source_url": "https://arxiv.org/abs/2609.40284", "catalog_id": "arxiv:2609.40284", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-30", "identifiers": { "arxiv": "2609.40284" }, "record_updated_at": "2026-09-30" }, "genre": "preprint", "id": "arxiv:2609.40284", "issued": { "date-parts": [ [ 2026, 9, 30 ] ] }, "note": "First published: 2026-09-30; Recorded arXiv version: v1; Recorded version updated: 2026-09-30T17:48:06Z.", "title": "cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.01491v1", "archive": "arXiv", "archive_location": "2610.01491", "author": [ { "literal": "Chengguang Gan" }, { "literal": "Zimeng He" }, { "literal": "Yoshihiro Tsujii" }, { "literal": "Ken-ichiro Kobayashi" }, { "literal": "Hiroki Itoh" }, { "literal": "Kotaro Funakoshi" } ], "citation-key": "arxiv_2610_01491", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9a5b97264b3928c8dc74b955cc4baa7005cd34b0ce9e42cbc48a63bc6dd1f5a7", "submitted_at": "2026-10-01T11:31:37Z", "updated_at": "2026-10-01T11:31:37Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.01491v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.01491", "catalog_id": "arxiv:2610.01491", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-01", "identifiers": { "arxiv": "2610.01491" }, "record_updated_at": "2026-10-01" }, "genre": "preprint", "id": "arxiv:2610.01491", "issued": { "date-parts": [ [ 2026, 10, 1 ] ] }, "note": "First published: 2026-10-01; Recorded arXiv version: v1; Recorded version updated: 2026-10-01T11:31:37Z.", "title": "Auditing Web Agent Evaluation on WebArena-Lite: Human Review of Outcomes and Trajectories", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.01787v1", "archive": "arXiv", "archive_location": "2610.01787", "author": [ { "literal": "Beining Wu" }, { "literal": "Zihao Ding" }, { "literal": "Jun Huang" } ], "citation-key": "arxiv_2610_01787", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7e85a601c31c6d06831ed1c590deef29d1c365e5d4751a308e42e8496ef11b19", "submitted_at": "2026-10-01T14:34:32Z", "updated_at": "2026-10-01T14:34:32Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.01787v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.01787", "catalog_id": "arxiv:2610.01787", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-01", "identifiers": { "arxiv": "2610.01787" }, "record_updated_at": "2026-10-01" }, "genre": "preprint", "id": "arxiv:2610.01787", "issued": { "date-parts": [ [ 2026, 10, 1 ] ] }, "note": "First published: 2026-10-01; Recorded arXiv version: v1; Recorded version updated: 2026-10-01T14:34:32Z.", "title": "Not All Experience Belongs in the Weights: Component Routing for Self-Improving GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.02542v1", "archive": "arXiv", "archive_location": "2610.02542", "author": [ { "literal": "Dhananjay Ashok" }, { "literal": "Shantanu Agarwal" }, { "literal": "Vivek Datla" }, { "literal": "Jonathan May" }, { "literal": "Alfy Samuel" } ], "citation-key": "arxiv_2610_02542", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "35ded32e7bac9a875a0fa41efad43fed5fb0aab54fade5cfc2a0628b1e666587", "submitted_at": "2026-10-01T22:30:08Z", "updated_at": "2026-10-01T22:30:08Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.02542v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.02542", "catalog_id": "arxiv:2610.02542", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-10-01", "identifiers": { "arxiv": "2610.02542" }, "record_updated_at": "2026-10-01" }, "genre": "preprint", "id": "arxiv:2610.02542", "issued": { "date-parts": [ [ 2026, 10, 1 ] ] }, "note": "First published: 2026-10-01; Recorded arXiv version: v1; Recorded version updated: 2026-10-01T22:30:08Z.", "title": "How To Train Your World Model: Fine-tuning vs RAG for LM-based World Modeling", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.02617v1", "archive": "arXiv", "archive_location": "2610.02617", "author": [ { "literal": "Yun-Yun Tsai" }, { "literal": "Yuning Mao" }, { "literal": "Shiqi Wang" }, { "literal": "Junfeng Yang" }, { "literal": "Sinong Wang" } ], "citation-key": "arxiv_2610_02617", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7ed117876ece99f55e241e6365d0e785b907f9a7a24cceb01b1fd757191a35c9", "submitted_at": "2026-10-02T00:17:11Z", "updated_at": "2026-10-02T00:17:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.02617v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.02617", "catalog_id": "arxiv:2610.02617", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-10-02", "identifiers": { "arxiv": "2610.02617" }, "record_updated_at": "2026-10-02" }, "genre": "preprint", "id": "arxiv:2610.02617", "issued": { "date-parts": [ [ 2026, 10, 2 ] ] }, "note": "First published: 2026-10-02; Recorded arXiv version: v1; Recorded version updated: 2026-10-02T00:17:11Z.", "title": "WebUIProof: Benchmarking WebUI Code Generators with UI-Agent Execution Harness", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.03089v1", "archive": "arXiv", "archive_location": "2610.03089", "author": [ { "literal": "Giulio Zingrillo" }, { "literal": "Hanna Foerster" }, { "literal": "Ilia Shumailov" }, { "literal": "Yiren Zhao" }, { "literal": "Robert Mullins" } ], "citation-key": "arxiv_2610_03089", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7469484bb1186dd633b26e41dddc890008e893c303f93d2352f93c9d47fd78c2", "submitted_at": "2026-10-02T10:08:39Z", "updated_at": "2026-10-02T10:08:39Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.03089v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.03089", "catalog_id": "arxiv:2610.03089", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-02", "identifiers": { "arxiv": "2610.03089" }, "record_updated_at": "2026-10-02" }, "genre": "preprint", "id": "arxiv:2610.03089", "issued": { "date-parts": [ [ 2026, 10, 2 ] ] }, "note": "First published: 2026-10-02; Recorded arXiv version: v1; Recorded version updated: 2026-10-02T10:08:39Z.", "title": "Securing Computer-Use Agents Against Branch Steering Attacks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.03818v2", "archive": "arXiv", "archive_location": "2610.03818", "author": [ { "literal": "Maria Sanchez del Rio" }, { "literal": "Mikhail Terekhov" } ], "citation-key": "arxiv_2610_03818", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f9dd28bdfb8879cbdb68f539c286365a06f4e9d04eba2420bb46f3575fdbfa05", "submitted_at": "2026-10-02T00:31:34Z", "updated_at": "2026-10-12T23:17:28Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2610.03818v2" }, "canonical_source_url": "https://arxiv.org/abs/2610.03818", "catalog_id": "arxiv:2610.03818", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-02", "identifiers": { "arxiv": "2610.03818" }, "record_updated_at": "2026-10-12" }, "genre": "preprint", "id": "arxiv:2610.03818", "issued": { "date-parts": [ [ 2026, 10, 2 ] ] }, "note": "First published: 2026-10-02; Recorded arXiv version: v2; Recorded version updated: 2026-10-12T23:17:28Z.", "title": "Control OSWorld: An AI Control Environment for GUI Computer Use Agents", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2610.04800v1", "archive": "arXiv", "archive_location": "2610.04800", "author": [ { "literal": "Ziyang Long" }, { "literal": "Xinqi Li" }, { "literal": "Lujing Xing" }, { "literal": "Hsin-Jung Yang" } ], "citation-key": "arxiv_2610_04800", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "baacd8b4fad7b146bb9152a54d60a617fdbed5d0142c0c0768dacea2e7f18e60", "submitted_at": "2026-10-03T22:56:09Z", "updated_at": "2026-10-03T22:56:09Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.04800v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.04800", "catalog_id": "arxiv:2610.04800", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-03", "identifiers": { "arxiv": "2610.04800" }, "record_updated_at": "2026-10-03" }, "genre": "preprint", "id": "arxiv:2610.04800", "issued": { "date-parts": [ [ 2026, 10, 3 ] ] }, "note": "First published: 2026-10-03; Recorded arXiv version: v1; Recorded version updated: 2026-10-03T22:56:09Z.", "title": "MedImageOSWorld: Benchmarking GUI Agents for Medical Image Consoles", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.06829v1", "archive": "arXiv", "archive_location": "2610.06829", "author": [ { "literal": "Yifan Zhang" }, { "literal": "Yutong Dai" }, { "literal": "Viraj Prabhu" }, { "literal": "Zhiyuan Hu" }, { "literal": "Ran Xu" }, { "literal": "Zeyuan Chen" } ], "citation-key": "arxiv_2610_06829", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "2f3e9860574951cd6b749d47da748e4aec36d4f5afe226f4a3d0d5459fd07935", "submitted_at": "2026-10-05T17:57:51Z", "updated_at": "2026-10-05T17:57:51Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.06829v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.06829", "catalog_id": "arxiv:2610.06829", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-05", "identifiers": { "arxiv": "2610.06829" }, "record_updated_at": "2026-10-05" }, "genre": "preprint", "id": "arxiv:2610.06829", "issued": { "date-parts": [ [ 2026, 10, 5 ] ] }, "note": "First published: 2026-10-05; Recorded arXiv version: v1; Recorded version updated: 2026-10-05T17:57:51Z.", "title": "CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.06898v1", "archive": "arXiv", "archive_location": "2610.06898", "author": [ { "literal": "Li Hu" }, { "literal": "Kanghua Mo" }, { "literal": "Yingbin Jin" }, { "literal": "Qingqing Ye" }, { "literal": "Haibo Hu" } ], "citation-key": "arxiv_2610_06898", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5afea9414462386ee921b9ead9381c0204963c710c9f616a96f4bc47079f059e", "submitted_at": "2026-09-29T03:03:00Z", "updated_at": "2026-09-29T03:03:00Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.06898v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.06898", "catalog_id": "arxiv:2610.06898", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-09-29", "identifiers": { "arxiv": "2610.06898" }, "record_updated_at": "2026-09-29" }, "genre": "preprint", "id": "arxiv:2610.06898", "issued": { "date-parts": [ [ 2026, 9, 29 ] ] }, "note": "First published: 2026-09-29; Recorded arXiv version: v1; Recorded version updated: 2026-09-29T03:03:00Z.", "title": "DIBench: Benchmarking Decision Integrity of GUI-based Mobile Agents Under Deceptive Injections", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.07444v1", "archive": "arXiv", "archive_location": "2610.07444", "author": [ { "literal": "Aadi Chauhan" }, { "literal": "Arthur Ilyasov" } ], "citation-key": "arxiv_2610_07444", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "81342daf25b0fb26f6481fcf836c3d958e26c2358be96eadb51e6914ea87e43c", "submitted_at": "2026-10-05T21:50:50Z", "updated_at": "2026-10-05T21:50:50Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.07444v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.07444", "catalog_id": "arxiv:2610.07444", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-05", "identifiers": { "arxiv": "2610.07444" }, "record_updated_at": "2026-10-05" }, "genre": "preprint", "id": "arxiv:2610.07444", "issued": { "date-parts": [ [ 2026, 10, 5 ] ] }, "note": "First published: 2026-10-05; Recorded arXiv version: v1; Recorded version updated: 2026-10-05T21:50:50Z.", "title": "Decoupling What from Where: How Should a Small GUI Grounding Model Receive the Action Type?", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.07972v1", "archive": "arXiv", "archive_location": "2610.07972", "author": [ { "literal": "Yeji Park" }, { "literal": "Jaeyun Shim" }, { "literal": "Taesik Gong" } ], "citation-key": "arxiv_2610_07972", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "b792cc0355a0947c61014c471c482aef173f761f7396a939c14816ac644abe66", "submitted_at": "2026-10-06T08:38:30Z", "updated_at": "2026-10-06T08:38:30Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.07972v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.07972", "catalog_id": "arxiv:2610.07972", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-06", "identifiers": { "arxiv": "2610.07972" }, "record_updated_at": "2026-10-06" }, "genre": "preprint", "id": "arxiv:2610.07972", "issued": { "date-parts": [ [ 2026, 10, 6 ] ] }, "note": "First published: 2026-10-06; Recorded arXiv version: v1; Recorded version updated: 2026-10-06T08:38:30Z.", "title": "Can Agents Work for Everyone? Cross-User Reliability for Mobile GUI Agents in Personalized User Interfaces", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.09240v1", "archive": "arXiv", "archive_location": "2610.09240", "author": [ { "literal": "Wanjing Han" }, { "literal": "Levi Taiji Li" }, { "literal": "Mu Zhang" }, { "literal": "Yue Jiang" }, { "literal": "Guanhong Tao" } ], "citation-key": "arxiv_2610_09240", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3b6ef7e785ad509ebd13fdb3da97fc583ee4c28dd42b4eab842c6cd88d7ef51f", "submitted_at": "2026-10-07T00:05:53Z", "updated_at": "2026-10-07T00:05:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.09240v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.09240", "catalog_id": "arxiv:2610.09240", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-07", "identifiers": { "arxiv": "2610.09240" }, "record_updated_at": "2026-10-07" }, "genre": "preprint", "id": "arxiv:2610.09240", "issued": { "date-parts": [ [ 2026, 10, 7 ] ] }, "note": "First published: 2026-10-07; Recorded arXiv version: v1; Recorded version updated: 2026-10-07T00:05:53Z.", "title": "Adversarial Images Hijack Web Agents from Visual Grounding to Browser Execution", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.09944v2", "archive": "arXiv", "archive_location": "2610.09944", "author": [ { "literal": "Michael Ofengenden" }, { "literal": "Maksym Andriushchenko" } ], "citation-key": "arxiv_2610_09944", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "03aea09f3ed1b75a8a6584efdfa51d45d1cfa9dc53fb3a25efd820c5b72e094a", "submitted_at": "2026-10-07T12:26:24Z", "updated_at": "2026-10-08T09:06:36Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2610.09944v2" }, "canonical_source_url": "https://arxiv.org/abs/2610.09944", "catalog_id": "arxiv:2610.09944", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-07", "identifiers": { "arxiv": "2610.09944" }, "record_updated_at": "2026-10-08" }, "genre": "preprint", "id": "arxiv:2610.09944", "issued": { "date-parts": [ [ 2026, 10, 7 ] ] }, "note": "First published: 2026-10-07; Recorded arXiv version: v2; Recorded version updated: 2026-10-08T09:06:36Z.", "title": "AgentTime: Can Agents Estimate and Control Their Own Runtime?", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2610.11050v1", "archive": "arXiv", "archive_location": "2610.11050", "author": [ { "literal": "Xing Han Lù" }, { "literal": "Dheeraj Vattikonda" }, { "literal": "Sina Hajimiri" }, { "literal": "Fatemeh Pesaran Zadeh" }, { "literal": "Parishad BehnamGhader" }, { "literal": "Ghazwa Darwiche" }, { "literal": "Amirhossein Kazemnejad" }, { "literal": "Christopher Pal" }, { "literal": "Alexandre Drouin" }, { "literal": "Siva Reddy" } ], "citation-key": "arxiv_2610_11050", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "37413f11bf8263f1516a8edca8c594227fcab104c0c767a7daa8eead9645ec34", "submitted_at": "2026-10-08T01:03:48Z", "updated_at": "2026-10-08T01:03:48Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.11050v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.11050", "catalog_id": "arxiv:2610.11050", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-08", "identifiers": { "arxiv": "2610.11050" }, "record_updated_at": "2026-10-08" }, "genre": "preprint", "id": "arxiv:2610.11050", "issued": { "date-parts": [ [ 2026, 10, 8 ] ] }, "note": "First published: 2026-10-08; Recorded arXiv version: v1; Recorded version updated: 2026-10-08T01:03:48Z.", "title": "AgentHorizon: Evaluating Agentic Judges for Long-Horizon Computer-Use Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.11942v1", "archive": "arXiv", "archive_location": "2610.11942", "author": [ { "literal": "Jiaming Zhang" }, { "literal": "Xuan Wang" }, { "literal": "Fuyao Zhang" }, { "literal": "Yang Cao" }, { "literal": "Lingjuan Lyu" }, { "literal": "Wei Yang Bryan Lim" } ], "citation-key": "arxiv_2610_11942", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "53120f960c7b98e87210c93ac661a2e397b858cc9f211786b0f8233fa61c6dd3", "submitted_at": "2026-10-08T13:33:23Z", "updated_at": "2026-10-08T13:33:23Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.11942v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.11942", "catalog_id": "arxiv:2610.11942", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-08", "identifiers": { "arxiv": "2610.11942" }, "record_updated_at": "2026-10-08" }, "genre": "preprint", "id": "arxiv:2610.11942", "issued": { "date-parts": [ [ 2026, 10, 8 ] ] }, "note": "First published: 2026-10-08; Recorded arXiv version: v1; Recorded version updated: 2026-10-08T13:33:23Z.", "title": "Right Screen, Wrong Transition: World Models as Verifiers for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.11963v1", "archive": "arXiv", "archive_location": "2610.11963", "author": [ { "literal": "Utku Boran Torun" }, { "literal": "Veli Karakaya" }, { "literal": "Eray Tüzün" } ], "citation-key": "arxiv_2610_11963", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0e77dcf101ba95684aac51084e43b914c0aca86fc25d061612224118004bdb21", "submitted_at": "2026-10-08T13:44:53Z", "updated_at": "2026-10-08T13:44:53Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.11963v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.11963", "catalog_id": "arxiv:2610.11963", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-08", "identifiers": { "arxiv": "2610.11963" }, "record_updated_at": "2026-10-08" }, "genre": "preprint", "id": "arxiv:2610.11963", "issued": { "date-parts": [ [ 2026, 10, 8 ] ] }, "note": "First published: 2026-10-08; Recorded arXiv version: v1; Recorded version updated: 2026-10-08T13:44:53Z.", "title": "Can LLMs Fix It Without Code? Toward Automated Verification of No-Code Bug Fixes", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.12611v2", "archive": "arXiv", "archive_location": "2610.12611", "author": [ { "literal": "Tian Jin" }, { "literal": "Wenhao Wang" }, { "literal": "Jing Kang" }, { "literal": "Haoting Shi" }, { "literal": "Rui Ye" }, { "literal": "Yijun Lu" }, { "literal": "Siheng Chen" } ], "citation-key": "arxiv_2610_12611", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "06b25b25e804683e11305e21dd04feb9131aad27f901998eda70ac91e527c6b9", "submitted_at": "2026-10-08T18:31:16Z", "updated_at": "2026-10-14T08:40:10Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2610.12611v2" }, "canonical_source_url": "https://arxiv.org/abs/2610.12611", "catalog_id": "arxiv:2610.12611", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-08", "identifiers": { "arxiv": "2610.12611" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.12611", "issued": { "date-parts": [ [ 2026, 10, 8 ] ] }, "note": "First published: 2026-10-08; Recorded arXiv version: v2; Recorded version updated: 2026-10-14T08:40:10Z.", "title": "SoftwareGym: Scaling Computer-Use Agent Training with Diverse Synthesized Software Environments", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2610.12823v1", "archive": "arXiv", "archive_location": "2610.12823", "author": [ { "literal": "Yisheng Li" }, { "literal": "Dongming Han" }, { "literal": "Hongkun Pan" }, { "literal": "Bingcheng Mao" }, { "literal": "Zijian Wang" }, { "literal": "Yunlong Mao" }, { "literal": "Wei Tong" }, { "literal": "Sheng Zhong" } ], "citation-key": "arxiv_2610_12823", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e6f1521fdfdf6a11a2a195feb05fd915afd6e9bb2332486b870da6375214ac1a", "submitted_at": "2026-10-09T00:03:13Z", "updated_at": "2026-10-09T00:03:13Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.12823v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.12823", "catalog_id": "arxiv:2610.12823", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-09", "identifiers": { "arxiv": "2610.12823" }, "record_updated_at": "2026-10-09" }, "genre": "preprint", "id": "arxiv:2610.12823", "issued": { "date-parts": [ [ 2026, 10, 9 ] ] }, "note": "First published: 2026-10-09; Recorded arXiv version: v1; Recorded version updated: 2026-10-09T00:03:13Z.", "title": "When Charts Deceive Web Agents: A Paired Benchmark for Misleading Visualization-Grounded Tasks", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.12866v1", "archive": "arXiv", "archive_location": "2610.12866", "author": [ { "literal": "Shangquan Sun" }, { "literal": "Ruoyu Chen" }, { "literal": "Zhirui Liu" }, { "literal": "Xiang Liu" }, { "literal": "Chenghao Qian" }, { "literal": "Wenqi Ren" }, { "literal": "Xiaochun Cao" } ], "citation-key": "arxiv_2610_12866", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7f7effcf5e101cfcc7e229fe4aab40d07401b1dd98cf5ce3b566fef863a42190", "submitted_at": "2026-10-09T01:50:26Z", "updated_at": "2026-10-09T01:50:26Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.12866v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.12866", "catalog_id": "arxiv:2610.12866", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-09", "identifiers": { "arxiv": "2610.12866" }, "record_updated_at": "2026-10-09" }, "genre": "preprint", "id": "arxiv:2610.12866", "issued": { "date-parts": [ [ 2026, 10, 9 ] ] }, "note": "First published: 2026-10-09; Recorded arXiv version: v1; Recorded version updated: 2026-10-09T01:50:26Z.", "title": "CaptchaAgent: Verifiable Visual Reasoning for CAPTCHA Solving", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.13092v1", "archive": "arXiv", "archive_location": "2610.13092", "author": [ { "literal": "Yuchen Ling" }, { "literal": "Shengcheng Yu" }, { "literal": "Zhenyu Chen" }, { "literal": "Chunrong Fang" } ], "citation-key": "arxiv_2610_13092", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "fb5f0220266a43b74e7a5f547abbe78a54a2ef993fee5d27f67827cb9715f5dc", "submitted_at": "2026-10-09T06:45:23Z", "updated_at": "2026-10-09T06:45:23Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.13092v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.13092", "catalog_id": "arxiv:2610.13092", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-09", "identifiers": { "arxiv": "2610.13092" }, "record_updated_at": "2026-10-09" }, "genre": "preprint", "id": "arxiv:2610.13092", "issued": { "date-parts": [ [ 2026, 10, 9 ] ] }, "note": "First published: 2026-10-09; Recorded arXiv version: v1; Recorded version updated: 2026-10-09T06:45:23Z.", "title": "When Faster Decisions Do Not Make Better Agents: An Empirical Study of Jev in Agent Workflows", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.13703v1", "archive": "arXiv", "archive_location": "2610.13703", "author": [ { "literal": "Vagner Figueredo de Santana" }, { "literal": "Diogo Cortiz" } ], "citation-key": "arxiv_2610_13703", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "447a4f283746069b2ea51d49c149379ad48a6e0b71f35dd15dedcb19417ab0cd", "submitted_at": "2026-10-09T14:59:24Z", "updated_at": "2026-10-09T14:59:24Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.13703v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.13703", "catalog_id": "arxiv:2610.13703", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-09", "identifiers": { "arxiv": "2610.13703" }, "record_updated_at": "2026-10-09" }, "genre": "preprint", "id": "arxiv:2610.13703", "issued": { "date-parts": [ [ 2026, 10, 9 ] ] }, "note": "First published: 2026-10-09; Recorded arXiv version: v1; Recorded version updated: 2026-10-09T14:59:24Z.", "title": "A Baseline Assessment for the Agentic Web Technology Stack", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.15138v1", "archive": "arXiv", "archive_location": "2610.15138", "author": [ { "literal": "Md Imran Hossen" } ], "citation-key": "arxiv_2610_15138", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "63e19265f1df2c336e28c4a13e11331021bd60866269c0b921d92a3c0fec080d", "submitted_at": "2026-10-10T19:30:31Z", "updated_at": "2026-10-10T19:30:31Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.15138v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.15138", "catalog_id": "arxiv:2610.15138", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-10", "identifiers": { "arxiv": "2610.15138" }, "record_updated_at": "2026-10-10" }, "genre": "preprint", "id": "arxiv:2610.15138", "issued": { "date-parts": [ [ 2026, 10, 10 ] ] }, "note": "First published: 2026-10-10; Recorded arXiv version: v1; Recorded version updated: 2026-10-10T19:30:31Z.", "title": "The Agent Clicks Approve: Injected Page Text Makes Browser Agents Authorize Access the Task Never Named", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.15738v1", "archive": "arXiv", "archive_location": "2610.15738", "author": [ { "literal": "Jianlin Chen" }, { "literal": "Wenhui Chen" }, { "literal": "Chi Man Vong" } ], "citation-key": "arxiv_2610_15738", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e6c60bb4d3b7bb2e2cba92201924d8da4f5e18ce5553c81fc9d5c6fb62b84751", "submitted_at": "2026-10-11T16:18:11Z", "updated_at": "2026-10-11T16:18:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.15738v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.15738", "catalog_id": "arxiv:2610.15738", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-11", "identifiers": { "arxiv": "2610.15738" }, "record_updated_at": "2026-10-11" }, "genre": "preprint", "id": "arxiv:2610.15738", "issued": { "date-parts": [ [ 2026, 10, 11 ] ] }, "note": "First published: 2026-10-11; Recorded arXiv version: v1; Recorded version updated: 2026-10-11T16:18:11Z.", "title": "Visible but Not Avoided: State-Dependent Re-Proposal of Rejected Candidates in Catalog-Based LLM Optimization Loops", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.16720v1", "archive": "arXiv", "archive_location": "2610.16720", "author": [ { "literal": "Jiahao Lu" }, { "literal": "Yize Shen" }, { "literal": "Chuanyuan Tan" }, { "literal": "Mianqiu Huang" }, { "literal": "Shicheng Fang" }, { "literal": "Xinchi Chen" }, { "literal": "Yining Zheng" }, { "literal": "Yuxin Wang" }, { "literal": "Xipeng Qiu" } ], "citation-key": "arxiv_2610_16720", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "73073d2d533307cd71b05afef2441a8a7b8f198f804f11f83536d4297db2491a", "submitted_at": "2026-10-12T12:52:21Z", "updated_at": "2026-10-12T12:52:21Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.16720v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.16720", "catalog_id": "arxiv:2610.16720", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-12", "identifiers": { "arxiv": "2610.16720" }, "record_updated_at": "2026-10-12" }, "genre": "preprint", "id": "arxiv:2610.16720", "issued": { "date-parts": [ [ 2026, 10, 12 ] ] }, "note": "First published: 2026-10-12; Recorded arXiv version: v1; Recorded version updated: 2026-10-12T12:52:21Z.", "title": "Learn from Move: the Next Step for GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.16778v1", "archive": "arXiv", "archive_location": "2610.16778", "author": [ { "literal": "Zian Yuan" }, { "literal": "Wei Li" } ], "citation-key": "arxiv_2610_16778", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "c5a2743a18bae1deeed13fbd1dd7379c3f21c67488627057ccb466ffe51d09dc", "submitted_at": "2026-10-12T13:34:52Z", "updated_at": "2026-10-12T13:34:52Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.16778v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.16778", "catalog_id": "arxiv:2610.16778", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-12", "identifiers": { "arxiv": "2610.16778" }, "record_updated_at": "2026-10-12" }, "genre": "preprint", "id": "arxiv:2610.16778", "issued": { "date-parts": [ [ 2026, 10, 12 ] ] }, "note": "First published: 2026-10-12; Recorded arXiv version: v1; Recorded version updated: 2026-10-12T13:34:52Z.", "title": "PrivacyGuard: Source-Linked Sanitization and Locally Resolvable PII References for Cloud GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.17457v1", "archive": "arXiv", "archive_location": "2610.17457", "author": [ { "literal": "Yuetian Chen" }, { "literal": "Kaiyuan Zhang" }, { "literal": "Yuntao Du" }, { "literal": "Thomas Li" }, { "literal": "Mark Tenenholtz" }, { "literal": "Kyle Polley" }, { "literal": "Jerry Ma" }, { "literal": "Denis Yarats" }, { "literal": "Ninghui Li" } ], "citation-key": "arxiv_2610_17457", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "247cd4c9634dea78780a2230138c728adf53871e8b65614697f76ba53dea858c", "submitted_at": "2026-10-12T22:17:51Z", "updated_at": "2026-10-12T22:17:51Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.17457v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.17457", "catalog_id": "arxiv:2610.17457", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-12", "identifiers": { "arxiv": "2610.17457" }, "record_updated_at": "2026-10-12" }, "genre": "preprint", "id": "arxiv:2610.17457", "issued": { "date-parts": [ [ 2026, 10, 12 ] ] }, "note": "First published: 2026-10-12; Recorded arXiv version: v1; Recorded version updated: 2026-10-12T22:17:51Z.", "title": "BrowseSafe++: Benchmarking Prompt Injection Detection Across Web Contexts, Languages, and Browser Views", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.17895v2", "archive": "arXiv", "archive_location": "2610.17895", "author": [ { "literal": "Wayne Chi" }, { "literal": "I-Hung Hsu" }, { "literal": "Li Chen" }, { "literal": "Wei-Lin Chiang" }, { "literal": "Ion Stoica" }, { "literal": "Anastasios Nikolas Angelopoulos" }, { "literal": "Ameet Talwalkar" }, { "literal": "Chris Donahue" }, { "literal": "Cho-Jui Hsieh" } ], "citation-key": "arxiv_2610_17895", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "14d2128155d1540955e54c3bdf9010724f27cf2e101079650315602856c98efa", "submitted_at": "2026-10-13T07:49:50Z", "updated_at": "2026-10-14T06:37:47Z", "version": 2, "versioned_url": "https://arxiv.org/abs/2610.17895v2" }, "canonical_source_url": "https://arxiv.org/abs/2610.17895", "catalog_id": "arxiv:2610.17895", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-13", "identifiers": { "arxiv": "2610.17895" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.17895", "issued": { "date-parts": [ [ 2026, 10, 13 ] ] }, "note": "First published: 2026-10-13; Recorded arXiv version: v2; Recorded version updated: 2026-10-14T06:37:47Z.", "title": "Interaction-as-a-Judge", "type": "article", "version": "2" }, { "URL": "https://arxiv.org/abs/2610.18067v1", "archive": "arXiv", "archive_location": "2610.18067", "author": [ { "literal": "Hang He" }, { "literal": "Linting Wang" }, { "literal": "Guanming Liu" }, { "literal": "Bo Mao" }, { "literal": "Jinxiu Liu" }, { "literal": "Yiran Suo" }, { "literal": "Maosen Zhou" }, { "literal": "Zhiheng Xi" }, { "literal": "Guanting Dong" }, { "literal": "Zhou Shao" }, { "literal": "Rui Zheng" }, { "literal": "Jie Zhou" }, { "literal": "Chengcheng Wan" }, { "literal": "Tao Gui" }, { "literal": "Qi Zhang" }, { "literal": "Xipeng Qiu" }, { "literal": "Xuanjing Huang" } ], "citation-key": "arxiv_2610_18067", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6add363f25f87ea1814c7b79ebf6b51f7431b3745a7d2d910960b78e2d97d13f", "submitted_at": "2026-10-13T09:52:02Z", "updated_at": "2026-10-13T09:52:02Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.18067v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.18067", "catalog_id": "arxiv:2610.18067", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-13", "identifiers": { "arxiv": "2610.18067" }, "record_updated_at": "2026-10-13" }, "genre": "preprint", "id": "arxiv:2610.18067", "issued": { "date-parts": [ [ 2026, 10, 13 ] ] }, "note": "First published: 2026-10-13; Recorded arXiv version: v1; Recorded version updated: 2026-10-13T09:52:02Z.", "title": "WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.18579v1", "archive": "arXiv", "archive_location": "2610.18579", "author": [ { "literal": "Yubo Li" }, { "literal": "Ramayya Krishnan" }, { "literal": "Rema Padman" } ], "citation-key": "arxiv_2610_18579", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "070f02f1ea798369817f471df5fe1d002b68d544135a1ac8e3a2aed2830e8c01", "submitted_at": "2026-10-13T15:43:30Z", "updated_at": "2026-10-13T15:43:30Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.18579v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.18579", "catalog_id": "arxiv:2610.18579", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-10-13", "identifiers": { "arxiv": "2610.18579" }, "record_updated_at": "2026-10-13" }, "genre": "preprint", "id": "arxiv:2610.18579", "issued": { "date-parts": [ [ 2026, 10, 13 ] ] }, "note": "First published: 2026-10-13; Recorded arXiv version: v1; Recorded version updated: 2026-10-13T15:43:30Z.", "title": "Reading Right, Acting Wrong: Diagnosing Decision Failures of System One Models in Agent Loops", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.19128v1", "archive": "arXiv", "archive_location": "2610.19128", "author": [ { "literal": "Yuxiang Lai" }, { "literal": "Jike Zhong" }, { "literal": "Chengzhi Liu" }, { "literal": "Ming Li" }, { "literal": "Haonan Wang" }, { "literal": "Xiaofeng Yang" }, { "literal": "Liang Zhao" }, { "literal": "Youbao Tang" }, { "literal": "Ruixi Fan" } ], "citation-key": "arxiv_2610_19128", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "5f72958f9a70e7714f17b0eb51c5453e08ea8ad4e8e7d4caa5e3170231088251", "submitted_at": "2026-10-13T20:01:18Z", "updated_at": "2026-10-13T20:01:18Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.19128v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.19128", "catalog_id": "arxiv:2610.19128", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-13", "identifiers": { "arxiv": "2610.19128" }, "record_updated_at": "2026-10-13" }, "genre": "preprint", "id": "arxiv:2610.19128", "issued": { "date-parts": [ [ 2026, 10, 13 ] ] }, "note": "First published: 2026-10-13; Recorded arXiv version: v1; Recorded version updated: 2026-10-13T20:01:18Z.", "title": "CLASHarness: Benchmarking Multi-Agent Conflict in Computer Use via a Controllable and Verifiable Generation Pipeline", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.19366v1", "archive": "arXiv", "archive_location": "2610.19366", "author": [ { "literal": "Minghao Guo" }, { "literal": "Qingyue Jiao" }, { "literal": "Xiujin Liu" }, { "literal": "Jianhong Pang" }, { "literal": "Ruimeng Yang" }, { "literal": "Naihao Xu" }, { "literal": "Liwei Che" }, { "literal": "Haochen Xue" }, { "literal": "Ziqi Wang" }, { "literal": "Xinying Cai" }, { "literal": "Jiaxuan Luo" }, { "literal": "Jiaxin Deng" }, { "literal": "Xuesong Ye" }, { "literal": "Jintao Qu" }, { "literal": "Wenyao Gao" }, { "literal": "Jingwen Gu" }, { "literal": "Chenyang Zhao" }, { "literal": "Shichun Liu" }, { "literal": "Guibin Zhang" }, { "literal": "Zefang Liu" }, { "literal": "Siyin Wang" }, { "literal": "Wujiang Xu" }, { "literal": "Danrui Li" }, { "literal": "Zeru Shi" }, { "literal": "Woongyeong Yeo" }, { "literal": "Shuang Liu" }, { "literal": "Wei-Chieh Huang" }, { "literal": "Weizhi Zhang" }, { "literal": "Tianyu Yang" }, { "literal": "Kangsan Kim" }, { "literal": "Yebowen Hu" }, { "literal": "Bowen Wen" }, { "literal": "Bowen Jiang" }, { "literal": "Haohao Su" }, { "literal": "Ming Yin" }, { "literal": "Sheldon Yu" }, { "literal": "Mian Zhang" }, { "literal": "Yuyan Chen" }, { "literal": "Hanrong Zhang" }, { "literal": "Kai Mei" }, { "literal": "Zihao Zhang" }, { "literal": "Mingyu Jin" }, { "literal": "Xi Zhu" }, { "literal": "Hengyi Wang" }, { "literal": "Tianxin Wei" }, { "literal": "Linyong Nan" }, { "literal": "Zhenwen Liang" }, { "literal": "Zhendong Chu" }, { "literal": "Ziqi Huang" }, { "literal": "Yongye Su" }, { "literal": "Sixun Dong" }, { "literal": "Shujian Liu" }, { "literal": "Kaiqu Liang" }, { "literal": "Jiaojiao Han" }, { "literal": "Zhaoye Fei" }, { "literal": "Kun Wang" }, { "literal": "Shihan Dou" }, { "literal": "Zhichun Guo" }, { "literal": "Xingjian Diao" }, { "literal": "Zeqi Gu" }, { "literal": "Heng Wang" }, { "literal": "Tao Feng" }, { "literal": "Cheng Qian" }, { "literal": "Hamid Palangi" }, { "literal": "Silei Xu" }, { "literal": "Xi Yun" }, { "literal": "Yi Zhu" }, { "literal": "Shilong Liu" }, { "literal": "Qingsong Wen" }, { "literal": "Sung Ju Hwang" }, { "literal": "Zuxuan Wu" }, { "literal": "Yu-Gang Jiang" }, { "literal": "Yiyu Shi" }, { "literal": "Dimitris N. Metaxas" }, { "literal": "Jiaxuan You" }, { "literal": "Julian McAuley" }, { "literal": "Mengdi Wang" }, { "literal": "Jiawei Han" }, { "literal": "Philip S. Yu" }, { "literal": "Ziyi Ye" } ], "citation-key": "arxiv_2610_19366", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7f1a48aa4b89fb97e8e1a186ba7345826a8b9efc9aa7b82e0dd1f5ce17d42b73", "submitted_at": "2026-10-14T01:11:57Z", "updated_at": "2026-10-14T01:11:57Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.19366v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.19366", "catalog_id": "arxiv:2610.19366", "catalog_kind": "survey", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.19366" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.19366", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T01:11:57Z.", "title": "Seeing, Maintaining, and Learning: The Evolution of Multimodal Agent Memory", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.19384v1", "archive": "arXiv", "archive_location": "2610.19384", "author": [ { "literal": "Yuke Wang" }, { "literal": "Benlei Cui" }, { "literal": "Shen Pang" }, { "literal": "Xuemei Dong" }, { "literal": "Longtao Huang" }, { "literal": "Hui Xue" }, { "literal": "Yuwen Zhai" }, { "literal": "Junjie Li" }, { "literal": "Jingqun Tang" }, { "literal": "Haiwen Hong" } ], "citation-key": "arxiv_2610_19384", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "da334d943d98b4bf9915c6ca28aab734d736f376e67cbab950aa1af65f540ebf", "submitted_at": "2026-10-14T01:30:03Z", "updated_at": "2026-10-14T01:30:03Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.19384v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.19384", "catalog_id": "arxiv:2610.19384", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.19384" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.19384", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T01:30:03Z.", "title": "MMA-SafetyBench: A Benchmark for Multimodal Agent Safety Evaluation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.19900v1", "archive": "arXiv", "archive_location": "2610.19900", "author": [ { "literal": "Ridong Zhang" }, { "literal": "Shuxian Liang" }, { "literal": "Chaoqun Wan" }, { "literal": "Zhen Huang" }, { "literal": "Xian-Sheng Hua" } ], "citation-key": "arxiv_2610_19900", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "50d5cb78e6715f0749c5973e900745ecdb3672febc2c8e090da94cb708127352", "submitted_at": "2026-10-14T09:58:40Z", "updated_at": "2026-10-14T09:58:40Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.19900v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.19900", "catalog_id": "arxiv:2610.19900", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.19900" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.19900", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T09:58:40Z.", "title": "How GUI Agents Fail: GraceBench for Execution Stability Diagnosis", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.20584v1", "archive": "arXiv", "archive_location": "2610.20584", "author": [ { "literal": "Mingyu Ouyang" }, { "literal": "Yiran Wang" }, { "literal": "Siyuan Hu" }, { "literal": "Xingyilang Yin" }, { "literal": "Huiqiang Sun" }, { "literal": "Guangzhi Wang" }, { "literal": "Junfu Pu" }, { "literal": "Cheng Cheng" }, { "literal": "Honghao Chen" }, { "literal": "Ying Shan" }, { "literal": "Mike Zheng Shou" }, { "literal": "Hwee Tou Ng" } ], "citation-key": "arxiv_2610_20584", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "67b539468bd2bbf0d3b49a5a7eed4791a8ddca4b56ad9515d23a29204d1f069f", "submitted_at": "2026-10-14T17:57:12Z", "updated_at": "2026-10-14T17:57:12Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.20584v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.20584", "catalog_id": "arxiv:2610.20584", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.20584" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.20584", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T17:57:12Z.", "title": "Player-1: Learning to Play from Synthetic Experience", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.20939v1", "archive": "arXiv", "archive_location": "2610.20939", "author": [ { "literal": "Panrong Tong" }, { "literal": "Quyu Kong" }, { "literal": "Pengxiang Li" }, { "literal": "Tianyu Xia" }, { "literal": "Hanzhang Zhou" }, { "literal": "Lei Wang" }, { "literal": "Chenglin Cai" }, { "literal": "Long Li" }, { "literal": "Long Chen" }, { "literal": "Jianan Zhang" }, { "literal": "Zhengwen Zeng" }, { "literal": "Xu Zhang" }, { "literal": "Gongjie Zhang" }, { "literal": "Yue Wang" }, { "literal": "Steven Hoi" } ], "citation-key": "arxiv_2610_20939", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "a5f4eda178718aac211eddc7281d572dc1d189dfd0cd130d4eba9f30f5a05ca2", "submitted_at": "2026-10-14T21:06:01Z", "updated_at": "2026-10-14T21:06:01Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.20939v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.20939", "catalog_id": "arxiv:2610.20939", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.20939" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.20939", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T21:06:01Z.", "title": "MAI-Gym: An Open All-in-One Platform for GUI Agent Evaluation and RL Training across Mobile and Computer Environments", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.22408v1", "archive": "arXiv", "archive_location": "2610.22408", "author": [ { "literal": "Tengda Guo" }, { "literal": "Jiawang Hu" }, { "literal": "Yifei Tao" }, { "literal": "Yin Wang" }, { "literal": "Haotian Hu" }, { "literal": "Fanyi Wang" }, { "literal": "Liujuan Tang" } ], "citation-key": "arxiv_2610_22408", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "025c7018c86ae4de79a659242070b2f91c5913333e9ecd4bd7fd357d13d27a34", "submitted_at": "2026-10-14T14:39:41Z", "updated_at": "2026-10-14T14:39:41Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.22408v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.22408", "catalog_id": "arxiv:2610.22408", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.22408" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.22408", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T14:39:41Z.", "title": "ECHOES: Seeing Beyond the Current Image via Recursive History Compression and Privileged On-Policy Distillation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.22672v1", "archive": "arXiv", "archive_location": "2610.22672", "author": [ { "literal": "Adamo Orsini" }, { "literal": "Seyed Mehdi Ayyoubzadeh" }, { "literal": "Sadman Rakib Pinon" }, { "literal": "Hannuo Zhang" }, { "literal": "Koosha Gholipour Baradari" }, { "literal": "Yuanhao Yu" }, { "literal": "Xinxin Zuo" }, { "literal": "Yang Wang" } ], "citation-key": "arxiv_2610_22672", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "caaf24ef4d8b86c8aaebc852235d0fe947a5ed859dc1ba8329ce07a2f7be9aa1", "submitted_at": "2026-10-15T21:40:34Z", "updated_at": "2026-10-15T21:40:34Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.22672v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.22672", "catalog_id": "arxiv:2610.22672", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-15", "identifiers": { "arxiv": "2610.22672" }, "record_updated_at": "2026-10-15" }, "genre": "preprint", "id": "arxiv:2610.22672", "issued": { "date-parts": [ [ 2026, 10, 15 ] ] }, "note": "First published: 2026-10-15; Recorded arXiv version: v1; Recorded version updated: 2026-10-15T21:40:34Z.", "title": "Not All Graphs Are Maps: A Task-Agnostic Exploration Framework for Mobile GUI Navigation", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.22990v1", "archive": "arXiv", "archive_location": "2610.22990", "author": [ { "literal": "Haoming Li" }, { "literal": "Defu Cao" }, { "literal": "Xu Hu" }, { "literal": "Yingtao Luo" }, { "literal": "Yiyang Wang" }, { "literal": "Xianghe Xu" }, { "literal": "Jiashu He" }, { "literal": "Wanting Jiang" }, { "literal": "Yifan Zhang" }, { "literal": "Yunhan Zhao" }, { "literal": "Siyang Wu" }, { "literal": "Chen Yang" }, { "literal": "Haixin Wang" }, { "literal": "Zheyuan Deng" }, { "literal": "Taoran Li" }, { "literal": "Jingwen Zhang" }, { "literal": "Zhejian Peng" }, { "literal": "Shushu Liang" }, { "literal": "Yue Zhao" }, { "literal": "Bingzhe Li" }, { "literal": "Feng Chen" }, { "literal": "Yifan Wang" }, { "literal": "Jintao Huang" }, { "literal": "Yijun Wang" }, { "literal": "Yuexing Hao" }, { "literal": "Xiaomin Li" } ], "citation-key": "arxiv_2610_22990", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "f613ef7ada260b494a411058bc733c7a5e0b074cdb01906c38132acf7275b9b1", "submitted_at": "2026-10-16T06:14:55Z", "updated_at": "2026-10-16T06:14:55Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.22990v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.22990", "catalog_id": "arxiv:2610.22990", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-16", "identifiers": { "arxiv": "2610.22990" }, "record_updated_at": "2026-10-16" }, "genre": "preprint", "id": "arxiv:2610.22990", "issued": { "date-parts": [ [ 2026, 10, 16 ] ] }, "note": "First published: 2026-10-16; Recorded arXiv version: v1; Recorded version updated: 2026-10-16T06:14:55Z.", "title": "PersonaCUABench: Benchmarking Computer-Use Agents Serving Diverse Personas", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.23303v1", "archive": "arXiv", "archive_location": "2610.23303", "author": [ { "literal": "Songping Wang" }, { "literal": "Borui Li" }, { "literal": "Qinglong Liu" }, { "literal": "Yueming Lyu" }, { "literal": "Yi Huang" }, { "literal": "Fengting Li" }, { "literal": "Chao Feng" }, { "literal": "Caifeng Shan" } ], "citation-key": "arxiv_2610_23303", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "ea753c1cb0666950d89f2781511f26872e3d5b1c2fd2434896248214160fa22c", "submitted_at": "2026-10-16T11:06:37Z", "updated_at": "2026-10-16T11:06:37Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.23303v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.23303", "catalog_id": "arxiv:2610.23303", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-16", "identifiers": { "arxiv": "2610.23303" }, "record_updated_at": "2026-10-16" }, "genre": "preprint", "id": "arxiv:2610.23303", "issued": { "date-parts": [ [ 2026, 10, 16 ] ] }, "note": "First published: 2026-10-16; Recorded arXiv version: v1; Recorded version updated: 2026-10-16T11:06:37Z.", "title": "MobileJailBench: Red Queen Jailbreak Benchmarking for Mobile Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.23307v1", "archive": "arXiv", "archive_location": "2610.23307", "author": [ { "literal": "Songping Wang" }, { "literal": "Peiyang Li" }, { "literal": "Yi Huang" }, { "literal": "Fengting Li" }, { "literal": "Xuhao Ran" }, { "literal": "Yueming Lyu" }, { "literal": "Chao Feng" }, { "literal": "Caifeng Shan" } ], "citation-key": "arxiv_2610_23307", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "3fc22fbeda19ac2947b00af73801f0674c550e0d7c78919f811cae754b06128c", "submitted_at": "2026-10-16T11:10:30Z", "updated_at": "2026-10-16T11:10:30Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.23307v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.23307", "catalog_id": "arxiv:2610.23307", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-16", "identifiers": { "arxiv": "2610.23307" }, "record_updated_at": "2026-10-16" }, "genre": "preprint", "id": "arxiv:2610.23307", "issued": { "date-parts": [ [ 2026, 10, 16 ] ] }, "note": "First published: 2026-10-16; Recorded arXiv version: v1; Recorded version updated: 2026-10-16T11:10:30Z.", "title": "MobileInject: Evaluating Reasoning-to-Action Security of MLLM-Based Mobile Agents under Cross-Modal Consistency Injection", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.23986v1", "archive": "arXiv", "archive_location": "2610.23986", "author": [ { "literal": "Xinyu Fang" }, { "literal": "Yuchi Wang" }, { "literal": "Weihao Yu" }, { "literal": "Rongyao Fang" }, { "literal": "Zhaohai Li" }, { "literal": "Zhixun Li" }, { "literal": "Tiancheng Gu" }, { "literal": "Xilin Wei" }, { "literal": "Yuhang Zang" }, { "literal": "Hongsheng Li" }, { "literal": "Dahua Lin" }, { "literal": "Guofeng Zhang" }, { "literal": "Shuai Bai" }, { "literal": "Zhibo Yang" }, { "literal": "Dayiheng Liu" } ], "citation-key": "arxiv_2610_23986", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "6dfa0c59ec1262cebcf2a4e61ab1e0b58c647513d8e6eedb2ce7e6a1a675558c", "submitted_at": "2026-10-14T12:46:54Z", "updated_at": "2026-10-14T12:46:54Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.23986v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.23986", "catalog_id": "arxiv:2610.23986", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "adjacent", "first_published_at": "2026-10-14", "identifiers": { "arxiv": "2610.23986" }, "record_updated_at": "2026-10-14" }, "genre": "preprint", "id": "arxiv:2610.23986", "issued": { "date-parts": [ [ 2026, 10, 14 ] ] }, "note": "First published: 2026-10-14; Recorded arXiv version: v1; Recorded version updated: 2026-10-14T12:46:54Z.", "title": "BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.24277v1", "archive": "arXiv", "archive_location": "2610.24277", "author": [ { "literal": "Egor Zverev" }, { "literal": "Juhee Kim" }, { "literal": "Leon Trampert" }, { "literal": "Boris Köpf" }, { "literal": "Aashish Kolluri" }, { "literal": "Shruti Tople" }, { "literal": "Santiago Zanella-Béguelin" } ], "citation-key": "arxiv_2610_24277", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "e675a3082f571da547fb5b6cb89e4f68a034bfa57236886e1218cb2bb2cdf1c3", "submitted_at": "2026-10-16T22:12:45Z", "updated_at": "2026-10-16T22:12:45Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.24277v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.24277", "catalog_id": "arxiv:2610.24277", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-16", "identifiers": { "arxiv": "2610.24277" }, "record_updated_at": "2026-10-16" }, "genre": "preprint", "id": "arxiv:2610.24277", "issued": { "date-parts": [ [ 2026, 10, 16 ] ] }, "note": "First published: 2026-10-16; Recorded arXiv version: v1; Recorded version updated: 2026-10-16T22:12:45Z.", "title": "SPIDES: Information-Flow Control for Browser Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.24372v1", "archive": "arXiv", "archive_location": "2610.24372", "author": [ { "literal": "Sizhe Tang" }, { "literal": "Hanhan Zhou" }, { "literal": "Jeffrey Svajlenko" }, { "literal": "Rashmi Gangadharaiah" } ], "citation-key": "arxiv_2610_24372", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9d6808cf26a609b50b7835ae3ecece0a13580aafcd589823f0b1eeff54732f69", "submitted_at": "2026-10-17T02:03:20Z", "updated_at": "2026-10-17T02:03:20Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.24372v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.24372", "catalog_id": "arxiv:2610.24372", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-17", "identifiers": { "arxiv": "2610.24372" }, "record_updated_at": "2026-10-17" }, "genre": "preprint", "id": "arxiv:2610.24372", "issued": { "date-parts": [ [ 2026, 10, 17 ] ] }, "note": "First published: 2026-10-17; Recorded arXiv version: v1; Recorded version updated: 2026-10-17T02:03:20Z.", "title": "Beyond Next-Screen Prediction: The Decision Value of GUI World Models", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.26071v1", "archive": "arXiv", "archive_location": "2610.26071", "author": [ { "literal": "Zhanhui Zhou" }, { "literal": "Weichen Zhang" }, { "literal": "Haoran Liu" }, { "literal": "Lingjie Chen" }, { "literal": "Tianneng Shi" }, { "literal": "Kevin Lin" }, { "literal": "Zhengyuan Yang" }, { "literal": "Lijuan Wang" }, { "literal": "Dawn Song" } ], "citation-key": "arxiv_2610_26071", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "0a297401f7ad4dcd2f485ade6bab295e365280bb803dcecc111aebf0693dd203", "submitted_at": "2026-10-19T09:30:54Z", "updated_at": "2026-10-19T09:30:54Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.26071v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.26071", "catalog_id": "arxiv:2610.26071", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-19", "identifiers": { "arxiv": "2610.26071" }, "record_updated_at": "2026-10-19" }, "genre": "preprint", "id": "arxiv:2610.26071", "issued": { "date-parts": [ [ 2026, 10, 19 ] ] }, "note": "First published: 2026-10-19; Recorded arXiv version: v1; Recorded version updated: 2026-10-19T09:30:54Z.", "title": "CUA-Lite: An Open Platform for Computer Use Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.26469v1", "archive": "arXiv", "archive_location": "2610.26469", "author": [ { "literal": "Zhizhou Wang" }, { "literal": "Yaxi Lu" }, { "literal": "Zhong Zhang" }, { "literal": "Zikun Hu" }, { "literal": "Xin Cong" }, { "literal": "Yankai Lin" }, { "literal": "Fangming Liu" } ], "citation-key": "arxiv_2610_26469", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "08ef7c9cdda652f60d3f2d4ab89e7218ad5a2b85f5c76f0277c289db06fd2796", "submitted_at": "2026-10-19T14:17:33Z", "updated_at": "2026-10-19T14:17:33Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.26469v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.26469", "catalog_id": "arxiv:2610.26469", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-19", "identifiers": { "arxiv": "2610.26469" }, "record_updated_at": "2026-10-19" }, "genre": "preprint", "id": "arxiv:2610.26469", "issued": { "date-parts": [ [ 2026, 10, 19 ] ] }, "note": "First published: 2026-10-19; Recorded arXiv version: v1; Recorded version updated: 2026-10-19T14:17:33Z.", "title": "VeriCUA: Learning Proactive Verification for Reliable GUI Agent Rewards", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.26503v1", "archive": "arXiv", "archive_location": "2610.26503", "author": [ { "literal": "Zhicheng Zheng" }, { "literal": "Cheng Zhang" }, { "literal": "Shengping Yu" }, { "literal": "Tianxiang Zhao" } ], "citation-key": "arxiv_2610_26503", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "7dc2ceb9aadbf7506efc068df99326fea5b60fdc6feb340c490841b88aa19c30", "submitted_at": "2026-10-19T14:38:41Z", "updated_at": "2026-10-19T14:38:41Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.26503v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.26503", "catalog_id": "arxiv:2610.26503", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-19", "identifiers": { "arxiv": "2610.26503" }, "record_updated_at": "2026-10-19" }, "genre": "preprint", "id": "arxiv:2610.26503", "issued": { "date-parts": [ [ 2026, 10, 19 ] ] }, "note": "First published: 2026-10-19; Recorded arXiv version: v1; Recorded version updated: 2026-10-19T14:38:41Z.", "title": "ScopeMem: Latent Memory That Learns Where to Revise Frozen GUI Agents", "type": "article", "version": "1" }, { "URL": "https://arxiv.org/abs/2610.26703v1", "archive": "arXiv", "archive_location": "2610.26703", "author": [ { "literal": "Ananya Gubbi Mohanbabu" }, { "literal": "Rosiana Natalie" }, { "literal": "Anhong Guo" }, { "literal": "Amy Pavel" } ], "citation-key": "arxiv_2610_26703", "custom": { "arxiv_snapshot": { "abstract_hash_method": "sha256-utf8-whitespace-collapsed-v1", "abstract_sha256": "9858b5ec15ee953504bd00e26b455c1e81104f80429c2dd112de17f9d23977ab", "submitted_at": "2026-10-19T16:39:11Z", "updated_at": "2026-10-19T16:39:11Z", "version": 1, "versioned_url": "https://arxiv.org/abs/2610.26703v1" }, "canonical_source_url": "https://arxiv.org/abs/2610.26703", "catalog_id": "arxiv:2610.26703", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "first_published_at": "2026-10-19", "identifiers": { "arxiv": "2610.26703" }, "record_updated_at": "2026-10-19" }, "genre": "preprint", "id": "arxiv:2610.26703", "issued": { "date-parts": [ [ 2026, 10, 19 ] ] }, "note": "First published: 2026-10-19; Recorded arXiv version: v1; Recorded version updated: 2026-10-19T16:39:11Z.", "title": "Comparing Computer Use Behavior Across Multiple Levels of Abstraction", "type": "article", "version": "1" }, { "URL": "https://proceedings.mlr.press/v70/shi17a.html", "author": [ { "literal": "Tianlin Shi" }, { "literal": "Andrej Karpathy" }, { "literal": "Linxi Fan" }, { "literal": "Jonathan Hernandez" }, { "literal": "Percy Liang" } ], "citation-key": "cua_ba263493b78b0263c7933138", "collection-title": "Proceedings of Machine Learning Research", "container-title": "Proceedings of the 34th International Conference on Machine Learning", "custom": { "canonical_source_url": "https://proceedings.mlr.press/v70/shi17a.html", "catalog_id": "url:52a3261eb6a19b402e88", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "publication", "first_published_at": "2017-07-17", "identifiers": {}, "publication": { "authors": [ "Tianlin Shi", "Andrej Karpathy", "Linxi Fan", "Jonathan Hernandez", "Percy Liang" ], "date_basis": "Publisher citation_publication_date metadata explicitly states 2017/07/17; this is separate from the conference interval.", "date_parts": [ 2017, 7, 17 ], "doi": null, "event_dates": "06--11 Aug (conference interval reported in the publisher BibTeX month field)", "evidence_urls": [ "https://proceedings.mlr.press/v70/shi17a.html", "https://proceedings.mlr.press/v70/shi17a/shi17a.pdf" ], "identity_basis": "No same-work record located in the full active catalog. Similarity screening was not used as automatic identity approval. The raw archive and exclusions are additionally checked during normalization.", "issue": null, "notes": "Publisher citation uses Tianlin Shi and Linxi Fan; the PDF also prints the nicknames Tim and Jim. The ordered publisher author names are retained. DOI not supplied in the retained publisher citation. Publisher metadata supplies the publication date separately from the conference interval.", "pages": "3135-3144", "pdf_url": "https://proceedings.mlr.press/v70/shi17a/shi17a.pdf", "publisher": "PMLR", "series": "Proceedings of Machine Learning Research", "title": "World of Bits: An Open-Domain Platform for Web-Based Agents", "type": "conference-paper", "url": "https://proceedings.mlr.press/v70/shi17a.html", "venue": "Proceedings of the 34th International Conference on Machine Learning", "volume": "70" }, "record_updated_at": null }, "id": "url:52a3261eb6a19b402e88", "issued": { "date-parts": [ [ 2017, 7, 17 ] ] }, "note": "Publication date: 2017-07-17.", "page": "3135-3144", "publisher": "PMLR", "title": "World of Bits: An Open-Domain Platform for Web-Based Agents", "type": "paper-conference", "volume": "70" }, { "DOI": "10.1145/1753326.1753555", "URL": "https://doi.org/10.1145/1753326.1753555", "author": [ { "literal": "Tsung-Hsiang Chang" }, { "literal": "Tom Yeh" }, { "literal": "Robert C. Miller" } ], "citation-key": "cua_08a6c6d3e0e8b509ecb90301", "container-title": "Proceedings of the 28th International Conference on Human Factors in Computing Systems (CHI '10)", "custom": { "canonical_source_url": "https://doi.org/10.1145/1753326.1753555", "catalog_id": "url:6262c757080f8838e62c", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "publication", "first_published_at": null, "identifiers": { "doi": "10.1145/1753326.1753555" }, "publication": { "authors": [ "Tsung-Hsiang Chang", "Tom Yeh", "Robert C. Miller" ], "date_basis": "Supplied month precision: 2010-04. Date Issued: April 2010", "date_parts": [ 2010, 4 ], "doi": "10.1145/1753326.1753555", "event_dates": null, "evidence_urls": [ "https://dspace.mit.edu/entities/publication/6a6097ba-1d97-43cf-bc70-04b5af02efd5" ], "identity_basis": "No same-work record located in the full active catalog. Similarity screening was not used as automatic identity approval. The raw archive and exclusions are additionally checked during normalization.", "issue": null, "notes": "Distinct visual-testing contribution from the 2009 Sikuli paper. The institutional abstract and published-version citation supply the retained evidence. Exact publication day and operating-system coverage are not established in the accessible institutional abstract.", "pages": "1535-1544", "pdf_url": null, "publisher": null, "series": null, "title": "GUI Testing Using Computer Vision", "type": "conference-paper", "url": "https://doi.org/10.1145/1753326.1753555", "venue": "Proceedings of the 28th International Conference on Human Factors in Computing Systems (CHI '10)", "volume": null }, "record_updated_at": null }, "id": "url:6262c757080f8838e62c", "issued": { "date-parts": [ [ 2010, 4 ] ] }, "note": "Publication date: 2010-04.", "page": "1535-1544", "title": "GUI Testing Using Computer Vision", "type": "paper-conference" }, { "URL": "https://openaccess.thecvf.com/content/CVPR2026/html/Tang_GUI-SAGE_Enhancing_GUI_Automation_with_Self-Explanatory_Learning_CVPR_2026_paper.html", "author": [ { "literal": "Fei Tang" }, { "literal": "Zhangxuan Gu" }, { "literal": "Zhengxi Lu" }, { "literal": "Shangzhan Zhang" }, { "literal": "Zhengwen Zeng" }, { "literal": "Shuheng Shen" }, { "literal": "Changhua Meng" }, { "literal": "Yuchen Yan" }, { "literal": "Wenqi Zhang" }, { "literal": "Yongliang Shen" }, { "literal": "Weiming Lu" }, { "literal": "Yueting Zhuang" } ], "citation-key": "cua_a6fbf231c6d3a11f67403503", "container-title": "Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)", "custom": { "canonical_source_url": "https://openaccess.thecvf.com/content/CVPR2026/html/Tang_GUI-SAGE_Enhancing_GUI_Automation_with_Self-Explanatory_Learning_CVPR_2026_paper.html", "catalog_id": "url:beb8f1fd6b945ffc9d93", "catalog_kind": "paper", "catalog_review_status": "curated", "catalog_scope": "core", "citation_edition": "publication", "first_published_at": null, "identifiers": {}, "publication": { "authors": [ "Fei Tang", "Zhangxuan Gu", "Zhengxi Lu", "Shangzhan Zhang", "Zhengwen Zeng", "Shuheng Shen", "Changhua Meng", "Yuchen Yan", "Wenqi Zhang", "Yongliang Shen", "Weiming Lu", "Yueting Zhuang" ], "date_basis": "Explicit publisher metadata: month = {June}, year = {2026}; precision=month.", "date_parts": [ 2026, 6 ], "doi": null, "event_dates": null, "evidence_urls": [ "https://openaccess.thecvf.com/content/CVPR2026/html/Tang_GUI-SAGE_Enhancing_GUI_Automation_with_Self-Explanatory_Learning_CVPR_2026_paper.html", "https://openaccess.thecvf.com/content/CVPR2026/papers/Tang_GUI-SAGE_Enhancing_GUI_Automation_with_Self-Explanatory_Learning_CVPR_2026_paper.pdf" ], "identity_basis": "No title/DOI/canonical URL match in the current paper catalog or raw archive; closest substantive same-team works were reread and have different contributions.", "issue": null, "notes": "CVF open-access accepted version; PDF states that the final proceedings version is on IEEE Xplore.", "pages": "13007-13016", "pdf_url": "https://openaccess.thecvf.com/content/CVPR2026/papers/Tang_GUI-SAGE_Enhancing_GUI_Automation_with_Self-Explanatory_Learning_CVPR_2026_paper.pdf", "publisher": null, "series": null, "title": "GUI-SAGE: Enhancing GUI Automation with Self-Explanatory Learning", "type": "conference-paper", "url": "https://openaccess.thecvf.com/content/CVPR2026/html/Tang_GUI-SAGE_Enhancing_GUI_Automation_with_Self-Explanatory_Learning_CVPR_2026_paper.html", "venue": "Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)", "volume": null }, "record_updated_at": null }, "id": "url:beb8f1fd6b945ffc9d93", "issued": { "date-parts": [ [ 2026, 6 ] ] }, "note": "Publication date: 2026-06.", "page": "13007-13016", "title": "GUI-SAGE: Enhancing GUI Automation with Self-Explanatory Learning", "type": "paper-conference" } ]