{
  "aliases": [],
  "authors": [],
  "details": {
    "benchmark_protocol": {
      "actions": [
        "list of UI element boxes [x1,y1,x2,y2] and text descriptions"
      ],
      "counts": [
        {
          "basis": "reported",
          "evidence_url": "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/README.md",
          "population": "English annotation subset reported by README",
          "qualifier": "exact",
          "unit": "page-annotation entries",
          "value": 1000
        },
        {
          "basis": "reported",
          "evidence_url": "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/README.md",
          "population": "Chinese annotation subset reported by README",
          "qualifier": "exact",
          "unit": "page-annotation entries",
          "value": 1000
        }
      ],
      "evaluation": "Metrics use greedy one-to-one matching in descending IoU order, then micro precision/recall/F1 and lowercased difflib.SequenceMatcher description similarity on matched pairs. Actual metrics.py accepts IoU >= 0.5 at its default threshold; README prose says > 0.5. Predictions align to annotations by image_path; callers must supply complete prediction coverage because missing predictions are not independently enumerated by evaluate_dataset.",
      "evidence_urls": [
        "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/README.md",
        "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/src/metrics.py",
        "https://api.github.com/repos/ant-research/GUIHalluBench/git/trees/master?recursive=1",
        "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/LEGAL.md",
        "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/.gitattributes"
      ],
      "focus": [
        "GUI page-element detection",
        "element-description consistency"
      ],
      "notes": "Each annotation entry represents a page, not one UI element. README’s illustrative num_samples=4796 output is not adopted as a corpus count. Public tree contains both annotation JSON files and an images.tar.gz Git-LFS pointer; payload availability/integrity was not tested by downloading images. LEGAL.md only governs comment-language precedence, not a code/data license grant; both slots remain unknown. No benchmark run was performed.",
      "observations": [
        "page screenshot"
      ],
      "setting": "offline"
    }
  },
  "environments": [
    "cross-platform"
  ],
  "evidence": [
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-16T06:40:51Z",
      "role": "primary",
      "sha256": "51df9dbcf52f46e4774207f2e6ad84dc8a6d117cdab6f77a38e9897ded1d0963",
      "supports": [
        "name",
        "kind",
        "summary",
        "environments",
        "topics",
        "scope",
        "relations"
      ],
      "url": "https://github.com/ant-research/GUIHalluBench"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-16T06:35:59Z",
      "role": "primary",
      "sha256": "d9c476d6960a4ef99e20d4c3b5d35a63f3ae5e61fc20e9dea00b11e64b74c9e9",
      "supports": [
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/README.md"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T14:21:09Z",
      "role": "primary",
      "sha256": "f36707dfcd3e83fc9d7f57b50d0c27007801e5029cc1ac255e000560ecf778fc",
      "supports": [
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/src/metrics.py"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T14:14:07Z",
      "role": "primary",
      "sha256": "c0896bc90d515b7bd8fe86b062aeb199bd47570e6eca8474986f5b1ef1d98dbd",
      "supports": [
        "details.benchmark_protocol"
      ],
      "url": "https://api.github.com/repos/ant-research/GUIHalluBench/git/trees/master?recursive=1"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T14:21:09Z",
      "role": "primary",
      "sha256": "adba50d9794b9ef3f7ec8cbc680f7f1fa3fbf9df0ac8d1f9b9ccab6d941bc11b",
      "supports": [
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/LEGAL.md"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T14:34:56Z",
      "role": "primary",
      "sha256": "8f35fa75be42a7a53e4f96de1b62be84962d2cf1beb8b23dfcfaecfaea4a6b90",
      "supports": [
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/ant-research/GUIHalluBench/HEAD/.gitattributes"
    }
  ],
  "id": "github:ant-research/guihallubench",
  "identifiers": {},
  "kind": "benchmark",
  "last_verified_at": "2026-10-16T06:40:51Z",
  "licenses": {
    "code": null,
    "data": null,
    "document": null,
    "weights": null
  },
  "links": [
    {
      "rel": "primary",
      "url": "https://github.com/ant-research/GUIHalluBench"
    }
  ],
  "name": "GUIHalluBench",
  "notes": "",
  "published_at": null,
  "relations": [],
  "review_status": "curated",
  "scope": "core",
  "summary": "Measures hallucination in UI-element parsing by matching the boxes and descriptions a model lists for 1,000 English and 1,000 Chinese screenshots against human annotations.",
  "topics": [
    "evaluation",
    "robustness"
  ],
  "updated_at": null,
  "url": "https://github.com/ant-research/GUIHalluBench"
}