{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.15064", "kind": "paper", "url": "https://arxiv.org/abs/2608.15064"}], "authors": ["Yuefeng Zou", "Yichen Lu", "Jingxiao Yang", "Bingtao Fu", "Gaoyang Zhang", "Xiongfei Bai", "Tian Chen", "Xiang Qi"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.15064", "source_url": "https://arxiv.org/abs/2608.15064"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "LongDocBench is a benchmark for Table-of-Contents Hierarchy Recovery and Contextual Relationship Recovery in long documents. It includes 85 real-world documents (financial reports, textbooks, academic papers) spanning 2,582 pages, with human-verified annotations for 3,937 heading nodes and 3,258 contextual relationships."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.15064", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.15064", "title": "LongDocBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Zhejiang University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.15064", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.15064", "source_url": "https://openalex.org/W7203674696"}, "key": "claire-radar:2608.15064", "modality": null, "name": "LongDocBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_longdocbench_5cb5bcf9", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "23eda4d3076e2636e89ddf9440b0caa556de04579d4d2cdb02142bf1e3a4a741", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.15064"}, "publisher": null, "released": "2026-08-15", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.15064", "source_url": "https://arxiv.org/abs/2608.15064"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-15064", "source": "claire_radar", "source_benchmark_id": "2608.15064", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.15064", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "bf1f43a4a7ff195e80e0cbf6f97d300f9114d2118aee8cde95428800039282d6", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "LongDocBench is a benchmark for Table-of-Contents Hierarchy Recovery and Contextual Relationship Recovery in long documents. It includes 85 real-world documents (financial reports, textbooks, academic papers) spanning 2,582 pages, with human-verified annotations for 3,937 heading nodes and 3,258 contextual relationships.", "domainCuration": {"method": "primary-source-review-v1", "state": "reviewed"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To benchmark these two tasks, we introduce \\textsc{LongDocBench}, comprising 85 real-world financial reports, textbooks, and academic papers spanning 2,582 pages, with up to 105 pages per document."}, "familyId": "bmf_25cf6240e611", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_longdocbench_5cb5bcf9", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.15064", "paper": "https://arxiv.org/abs/2608.15064", "pdf": "https://arxiv.org/pdf/2608.15064", "project": null, "report": "https://arxiv.org/abs/2608.15064"}, "metrics": [], "motivation": "Parsing visual documents into machine-readable representations is fundamental to document intelligence.", "name": "LongDocBench", "oneLine": "LongDocBench is a benchmark for Table-of-Contents Hierarchy Recovery and Contextual Relationship Recovery in long documents. It includes 85 real-world documents (financial reports, textbooks, academic papers) spanning 2,582 pages, with human-verified annotations for 3,937 heading nodes and 3,258 contextual relationships.", "paperTitle": "LongDocBench: Benchmarking TOC Hierarchy and Contextual Relationship Recovery in Long Documents", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.15064", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "LongDocBench Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2608.15064"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-15", "source": {"authors": ["Yuefeng Zou", "Yichen Lu", "Jingxiao Yang", "Bingtao Fu", "Gaoyang Zhang", "Xiongfei Bai", "Tian Chen", "Xiang Qi"], "categories": ["cs.AI"], "id": "2608.15064", "title": "LongDocBench: Benchmarking TOC Hierarchy and Contextual Relationship Recovery in Long Documents", "type": "arxiv", "url": "https://arxiv.org/abs/2608.15064"}, "sourceUpdatedAt": "2026-08-15T06:25:26Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Existing document parsing benchmarks focus on page-level tasks, leaving document-level structure recovery unevaluated. LongDocBench provides a standardized evaluation for tasks that are critical for understanding long documents, enabling comparison of parsers on hierarchy and relationship recovery."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
