{
  "added_at": "2026-11-01",
  "arxiv_id": "2505.20139",
  "arxiv_version_reviewed": "v3",
  "authors": [],
  "authors_omitted": true,
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-structeval.bib",
  "code_url": "https://github.com/TIGER-AI-Lab/StructEval",
  "date_basis": "Original arXiv first-submission date; reviewed preprint edition retained.",
  "date_basis_zh": "采用 arXiv 首次提交日期，并保留本次核验的预印本版本。",
  "date_precision": "day",
  "description_en": "Evaluates generation and conversion of structured outputs, including HTML, React and SVG, with separate checks for format adherence and structural correctness.",
  "description_zh": "评测结构化输出的生成与转换，包括 HTML、React 和 SVG，并分别检查格式遵循和结构正确性。",
  "doi": null,
  "entry_url": "https://awesomerea.org/papers/paper-structeval/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-structeval/",
  "evidence_urls": [
    "https://arxiv.org/abs/2505.20139v3",
    "https://github.com/TIGER-AI-Lab/StructEval",
    "https://tiger-ai-lab.github.io/StructEval/"
  ],
  "first_arxiv_date": "2025-05-26",
  "id": "paper-structeval",
  "json_url": "https://awesomerea.org/papers/paper-structeval/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-structeval/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-structeval/index.md",
  "metadata_notes": "This record describes the original arXiv edition. Supporting method.",
  "metadata_notes_zh": "本条描述原始 arXiv 版本。属于配套方法。",
  "pdf_url": "https://arxiv.org/pdf/2505.20139v3",
  "project_url": "https://tiger-ai-lab.github.io/StructEval/",
  "publication_date": "2025-05-26",
  "publication_status": "preprint",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "Not a software reverse-engineering or GUI-navigation benchmark. Structured-output scores do not prove recovered source identity, backend reconstruction or behavioral equivalence; aggregate scores are not fully correct-output percentages.",
  "scope_limitations_zh": "并非软件逆向工程或 GUI 导航基准；结构化输出得分不证明源码身份、后端恢复或行为等价，综合得分也不是完全正确输出的百分比。",
  "targets": [
    "website"
  ],
  "title": "StructEval: Benchmarking LLMs' Capabilities to Generate Structural Outputs",
  "title_zh": "StructEval：评测大语言模型生成结构化输出的能力",
  "url": "https://arxiv.org/abs/2505.20139",
  "venue": "arXiv",
  "verification": "source-reviewed",
  "verified_at": "2026-10-20",
  "why_fits_en": "A supporting method for checking structured and rendered artifacts produced during frontend reconstruction; parseability alone does not establish correct content.",
  "why_fits_zh": "可辅助检查前端重建产出的结构化与渲染工件；能够解析并不代表内容正确。",
  "year": 2025
}