Focused Omni-Modal GUI Agents
FocusGUI learns step-specific weights over screenshot, audio, video and interaction-context channels, injecting a compact residual into a frozen decoder without pruning its inputs.
Android
GUI 定位
已审阅并标注「GUI 定位」的研究与资产。
条目依据已审阅标签归类，同一主题不表示资源之间存在依赖关系。日期保留来源精度，注释保留原语言。
STP-GUI: Spatially-Aware Token Pruning for Efficient Web-GUI Agents
Introduces STP-GUI, which selects spatially coherent screenshot windows conditioned on the instruction to reduce visual tokens while retaining useful interface structure.
网页Android
Why Generate a Click? Plug-in Decision Heads for Web Agents
Adds decision heads to a frozen vision-language generator for selecting web actions and targets, with a confidence gate that falls back to text generation.
网页
MMA-SafetyBench: A Benchmark for Multimodal Agent Safety Evaluation
Builds trajectory-guided visual-injection tests for web, desktop and mobile observations, separating target-matching action predictions from broader semantic compromise.
网页桌面
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning
Introduces ScreenRef screenshot grounding data and a training sequence from supervised learning through teacher-guided distillation to reinforcement learning for GroundGUI-8B.
桌面网页跨平台
MHVF: Training-Free GUI Grounding via Verification-Guided Multi-Hypothesis Search
Introduces MHVF, a training-free grounding method that explores several candidate screen regions, verifies them in local views, and uses those checks to guide further search.
网页
Learn from Move: the Next Step for GUI Agents
Introduces LatentGUIWorld to test learning interface semantics and dynamics through interaction, and studies finer movement actions with intermediate visual feedback and exploration demonstrations.
Seg2Click: Learning Pixel-Level Action Masks for GUI Visual Grounding
Recasts GUI grounding as predicting a mask of valid click locations from bounding-box supervision, then derives a click point with a deterministic selection rule.
Visible but Not Avoided: State-Dependent Re-Proposal of Rejected Candidates in Catalog-Based LLM Optimization Loops
Studies repeated rejected edits while optimizing a screenshot-only GUI agent, contrasting static probes with reached loop states and testing how rejection wording changes subsequent edit proposals.
ConClick: Element-Wise Contrastive Learning Improves GUI Grounding
Analyzes recurring distractor clicks in GUI grounding and introduces ConClick, a contrastive training method built from target–distractor pairs. Its ReClick inference procedure revisits candidate regions and compares competing predictions.
CaptchaAgent: Verifiable Visual Reasoning for CAPTCHA Solving
Presents CaptchaAgent, which builds an explicit visual representation of objects and relations, refines ambiguous regions, and checks counterfactual consistency before acting on interactive CAPTCHAs.
网页
Adversarial Images Hijack Web Agents from Visual Grounding to Browser Execution
Introduces WebMirage, a visual red-teaming framework that models webpage element competition and action post-processing to test whether manipulated grounding decisions lead to executed browser actions.
网页
Decoupling What from Where: How Should a Small GUI Grounding Model Receive the Action Type?
Compares ways to condition a small GUI grounding model on action type under matched training conditions on Android in the Wild. The study traces apparent gains to a coordinate-serialization choice and reports limited evidence that the gains persist with predicted action types.
Android
Recurrent Latent Visual Search for GUI Grounding
Introduces ReLaViS, which performs recurrent visual search inside a grounding model by repeatedly querying screenshot tokens and feeding the resulting spatial evidence back into the model. Coarse-to-fine supervision is constructed from existing interface-element annotations.
DeskForge: Dense Supervision from Desktop Environments for Computer-Use Agents
Generates dense desktop supervision by varying real application scenes and combining screenshots, accessibility trees, and window geometry.
桌面
AutoGUIWorld: Image Generators as Visual World Models for GUI Agent
Introduces AutoGUIWorld, which combines task planning and image editing to synthesize action-grounded GUI trajectories without running the depicted software. It filters generated transitions and tests whether training on them improves agents operating real desktop and scientific applications.
桌面LinuxWindowsmacOS网页
SCA: Spatial Credit Assignment for Reinforcement Learning of GUI Agents
Introduces Spatial Credit Assignment, which uses click coordinates to refine group-relative reinforcement-learning updates for GUI grounding. It distinguishes failed clicks by spatial evidence while leaving the deployed action policy unchanged.
Structured Interaction, Visual Localization, and Robust Execution for Complex Web Tasks: A Technical Report on the WebRetriever Challenge
Describes a WebRetriever agent that combines semantic webpage operations with visual localization when structured representations are insufficient. Hierarchical context management and fault-aware execution support processing tasks across concurrent browsers.
网页
Probe to Act: Elevating Browser-Use Agent via Active Visual Probing
Introduces Probe to Act, which lets browser agents request visual evidence linking DOM handles and screen regions before committing an operation. It retains selected interaction evidence as memory and studies both prompted use and distillation into open models.
网页
DashAct: A Progressive Diagnostic Benchmark for GUI Agents in Interactive Dashboard Analysis
Diagnoses dashboard-agent failures by progressively restoring interaction context, next-action support, and visual target information within the same task.
ScreenHaystack: Finding Blind Zones in GUI Grounding
Moves controlled target icons across high-resolution interfaces to diagnose spatial blind zones in grounding models and study targeted augmentation.
跨平台
CaptchaArena: A Large-Scale, Fine-Grained Dataset for Training Computer-Use Agents on Interactive CAPTCHAs
Introduces CaptchaArena, a dataset of execution-verified interactive CAPTCHA solutions with screenshot–action trajectories, reasoning annotations, and pixel masks for irregular targets. The work trains a single policy across CAPTCHA types using supervised learning followed by verifier-reward reinforcement learning.
Guides That Cause Actions: An Offline Study of Guide-Action Mutual Reinforcement in Multimodal Web Agents
WebMRE pairs guidance sentences with grounded actions in an offline benchmark derived from successful WebArena traces. Controlled decoding-prefix interventions test how guide content affects action selection, while separating offline prediction metrics from live browser task success.
网页
How Many Pixels Is a Digit Worth? Place-Aware Coordinate Entropy for GUI Agent Confidence Estimation
PACE weights coordinate-digit entropy by place value to estimate confidence in GUI clicks from one forward pass. It evaluates correctness ranking and selective accuracy against repeated-sampling alternatives on ScreenSpot-Pro and ScreenSpot-v2.
MintAct: A Unified Visual Agent for Digital Environments
MintAct trains a visual-agent family for grounding, navigation and tool use across mobile, desktop and web environments. Its shared infrastructure supports trajectory collection and asynchronous online RL while controlling cross-domain sampling and off-policy drift.
跨平台
RankGround: Efficient High-Resolution GUI Grounding via Lightweight Reranker-Guided Crop Selection
RankGround trains a lightweight reranker to select a promising screenshot crop before one VLM localization call. Ranking supervision uses containment and boundary-aware augmentation, with pointwise and listwise training stages.
EchoPath: Execution-Level Replayable Memory for GUI Agents
EchoPath turns validated GUI trajectories into parameterized replay procedures with explicit runtime preconditions. Image-based target matching adjusts stored coordinates, while ambiguous steps fall back to bounded repair or fresh planning.
Learning from Reliable Negatives: Confidence-Anchored Test-Time Adaptation for GUI Grounding
Confidence-Anchored Learning uses coordinate-token confidence to filter pseudo-supervision for GUI grounding. Its negative-only variant performs label-free test-time updates to reduce reliance on potentially incorrect positive examples.
LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents
LLaDA-UI adapts block-wise diffusion generation to a vision-language GUI agent through multimodal pretraining and GUI supervised fine-tuning. It evaluates grounding and navigation across multiple platform settings.
网页桌面
Are We There Yet? Assessing Computer-Use Agents for Blind Users' Accessible Interaction with Desktop Applications
A diary study deploys the screen-reader-accessible OLLA computer-use prototype with blind users across desktop applications. Recorded commands and interaction traces support model comparisons and analysis of grounding, planning, constraint tracking and termination failures.
桌面
Iron: Intent-Aligned and Retrospective Dual Learning Framework for Enhancing Generalist Virtual Agents
Introduces Iron, a GUI-agent training framework that aligns low-level actions with high-level intent using cycle-consistency rewards. It also repurposes failed trajectories through hindsight reconstruction and evaluates transfer to other digital tasks.
网页
GUI-Primitives: Diagnosing Spatial Reasoning Failures in Vision-Language GUI Grounding
Introduces GUI-Primitives, a paired-instruction benchmark that changes a spatial relation while holding the screenshot and anchor fixed. Its diagnostics separate failure to localize either candidate from failure to choose the candidate satisfying the relation.
Lexical Coupling in GUI Element Grounding: Sentence Embeddings Track Labels across Mobile and Web
Studies whether text-embedding GUI grounding scores reflect semantic matching or recovery of visible labels. Same-screen ranking comparisons with lexical baselines expose differences between deployable fusion, oracle selection, and performance on label-poor targets.
网页跨平台
LookAgain: Closed-Loop GUI Grounding with Visually Grounded Reflection
LookAgain treats a GUI coordinate as a revisable hypothesis, rendering its marker and a local screenshot patch before confirmation. Reflective supervised trajectories provide a cold start, followed by GRPO using terminal grounding correctness.
Hallucination-Free GUI Grounding via Regression-Free Layout-Aware Matching
This framework separates frozen-model instruction expansion from a trained layout-aware matcher that localizes GUI elements without coordinate regression. Binary text/icon supervision supports the matcher, with evaluations on GUI grounding and Mind2Web element-selection tasks.
网页
MissClick: Execution-Aware Adversarial Attacks on Coordinate Generation in GUI Grounding Models
MissClick designs white-box attacks around the numerical effect of coordinate digits on executable GUI clicks. Separate untargeted displacement and targeted place-weighted objectives are evaluated across desktop, web and mobile grounding settings.
跨平台
GUI-Lens: Coarse-to-Fine Cropping for GUI Grounding with General-Purpose VLMs
GUI-Lens uses OCR and detected interface components as coordinate references while a general-purpose VLM selects progressively enlarged screenshot crops. Instruction-conditioned checks constrain the crop and click proposals before mapping the final location back to the original screen.
MAGA: Multi-Platform Self-Fusion of GUI Agents via Structured Action Distillation
MAGA distills specialized GUI agents into one policy by allocating supervision according to structured-action correctness rather than treating all response tokens equally. A training-only teacher hint supports fusion across mobile, web and desktop settings without changing student inputs.
跨平台
Tactile: Giving Computer-Using Agents Hands and Feet
Tactile turns accessibility, OCR and visual evidence into explicit desktop targets with executable actions and verification cues. Its execution loop chooses semantic actions or coordinate fallbacks while retaining provenance for replay and diagnosis.
桌面macOS
MAG: A Web-Agent Benchmark and Harness for Multimodal Action and Guide Generation
MAG jointly evaluates browser task actions and user-facing guides grounded in screenshots through element marks or pixel coordinates. Its harness covers annotation, live evaluation and joint metrics, with an expert-trajectory-augmented GRPO training study.
网页
DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching
DigitalCoach records expert–novice software coaching conversations aligned with screen and input events. Automated and interactive evaluations examine how model coaches differ from humans in explanation, error diagnosis and visual grounding when helping a learner operate software.
相关研究
LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents
LUMOS converts native accessibility metadata and browser UI structure into semantic interface blueprints with stable element identifiers and available actions. It combines live pointer-to-element grounding with an observe-act loop constrained to visible-UI primitives.
网页
One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding
InnerZoom preserves target-region cues from intermediate decoder layers and reinjects a compact evidence state during coordinate decoding. Its single-forward GUI grounder is compared with crop-and-rerun zooming and matched supervised/RL baselines for accuracy and inference cost.
GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots
GUICrafter first learns grounding from large collections of unannotated screenshots and webpages, then uses a smaller high-quality dataset for reinforcement-learning calibration. The curriculum targets reduced dependence on manually annotated GUI training data.
WinDOM: Self-Family Distillation for Small-Model GUI Grounding
WinDOM obtains bounding-box supervision from a Windows 11 web reimplementation driven through Playwright, and studies rejection-sampling distillation for small GUI grounders. It treats the depth of the supervised cold start as a GRPO choice, comparing same-size EMA and larger same-family teachers.
网页
Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets
Argus benchmarks post-hoc uncertainty methods for single-step GUI grounding across open-weight and closed-source observation regimes. It tests calibration, rejection, error severity and conformal click regions, finding that uncertainty rankings transfer less reliably across model/interface regimes than across datasets for one model.
Trust the Right Teacher: Quality-Aware Self-Distillation for GUI Grounding
Proposes quality-aware self-distillation for GUI coordinate generation under student-produced prefixes. A correctness-aware gate suppresses coordinate-token supervision that cannot reach the target box, while teacher-probability scaling adjusts the remaining signal strength.
VISTA: View-Consistent Self-Verified Training for GUI Grounding
VISTA forms GRPO comparison groups from target-preserving crops of the same GUI instance with exactly remapped target boxes. A conditional, advantage-weighted cross-view anchor supplies additional supervision without entering the group baseline, and evaluation includes view-consistency robustness.
Naive Visual Memory is Not Enough: A Failure-Mode Study of GUI Agents
This study finds that full-image visual memory can reduce some state errors while worsening action-level failures. Its AGMem alternative stores local crops associated with successful actions or recovery and evaluates their use in desktop GUI tasks.
桌面
A History-Aware Visually Grounded Critic for Computer Use Agents
HiViG combines a compact history of completed actions with a visually grounded critic that checks proposed coordinates before execution. Its evaluations separate history-related planning failures from errors caused by targeting the wrong screen element.
网页桌面
GUI-AC: Enhancing Continual Learning in GUI Agents
GUI-AC uses grounding certainty to adjust advantage weighting and policy-clipping bounds during continual reinforcement fine-tuning. Its mechanisms aim to suppress noisy confidence updates while retaining exploration under shifting GUI domains and resolutions.
RunAgent SuperBrowser: A Theory of Autonomous Web Navigation Grounded in Human Browsing Behaviour
SUPERBROWSER combines screenshot-based target proposals with separate orchestration, planning and action roles. A bounded ledger and context-eviction process retain recent actions, task facts and checkpoints during browser execution.
网页
DragOn: A Benchmark and Dataset for Drag-Based GUI Interactions
DragOn supplies drag-grounding training data and a held-out benchmark for highlighting text, selecting cells, resizing elements and moving sliders. It evaluates existing vision-language models and a model fine-tuned on the drag tasks.
DocOS: Towards Proactive Document-Guided Actions in GUI Agents
DocOS benchmarks agents that search online documentation and translate procedural instructions into GUI actions in interactive web environments. It separates failures to retrieve useful procedures from failures to ground retrieved instructions into executable actions.
网页
MUIAnno: An Expert-Annotated Dataset and Evaluation Benchmark for Mobile UI Understanding
MUIAnno provides expert labels for elements in manually explored mobile-app screens, with a web annotation tool that emits structured JSON. It also supplies UI-element detection baselines for the dataset.
SE-GA: Memory-Augmented Self-Evolution for GUI Agents
SE-GA retrieves hierarchical memories during inference and uses the collected experience in a separate policy-improvement pipeline. Its evaluation includes offline grounding/action tasks and online Android interaction, which measure different capabilities.
Android
DRS-GUI: Dynamic Region Search for Training-Free GUI Grounding
DRS-GUI searches screenshot regions through focus, shift and scatter operations selected by a tree-search planner. A region-quality reward guides this training-free perceptual search before final GUI grounding.
WinDeskGround: A Benchmark for Robust GUI Grounding in Complex Multi-Window Desktop Environments
WinDeskGround synthesizes multi-window desktop layouts with controlled occlusion, density and semantic similarity to test GUI grounding robustness. Its instruction-target benchmark measures how localization changes when interfaces depart from simple single-layer scenes.
桌面
Covering Human Action Space for Computer Use: Data Synthesis and Benchmark
Introduces CUActSpot to evaluate clicks, drags, drawing and other screen interactions across GUI, text, tables, canvases and natural images. A renderer-based pipeline records geometry and screenshots, then generates matching instructions and action traces for training a grounding model.
What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs
Re-Prefill adds an attention-guided second prefill pass that revisits candidate screen regions before coordinate generation. The study evaluates this training-free intervention across GUI-grounding models and benchmarks.
第 1–60 条，共 380 条资源。
通过主题导出清单获取完整记录和证据。克隆目录后，可将主题记录直接交给标准 RAG 与引用导出工具。