{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.07401", "kind": "paper", "url": "https://arxiv.org/abs/2606.07401"}], "authors": ["Ameya Joshi", "Joon Kim", "Gus Eggert", "Joseph Bajor", "Cindy Hao", "Jing Reyhan", "Kushal Byatnal", "Eli Badgio"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.07401", "source_url": "https://arxiv.org/abs/2606.07401"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "RealDocBench evaluates field-level QA and layout understanding on real regulated documents, with 1,356 field-level questions over 581 documents and 1,500 annotated page images, scored on per-field accuracy and adjacency-aware layout metrics."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.07401", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.07401", "title": "RealDocBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.07401", "modality": null, "name": "RealDocBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "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_realdocbench_6da01042", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "2352a8f5655dd29f725f9d747a31fb3b898202d66e775cf0e1e0cf1ec5f27ac1", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.07401"}, "publisher": null, "released": "2026-06-05", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.07401", "source_url": "https://arxiv.org/abs/2606.07401"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-07401", "source": "claire_radar", "source_benchmark_id": "2606.07401", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Machine Generated", "applicationDomains": ["General AI"], "area": "Vision & 3D", "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/2606.07401", "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": "Medium", "construction": "Original Synthetic", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "a0e2f39a9d4bf85b28d3b984ddccb01bba575c8d2a795f922bae59dab2c23e51", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "RealDocBench evaluates field-level QA and layout understanding on real regulated documents, with 1,356 field-level questions over 581 documents and 1,500 annotated page images, scored on per-field accuracy and adjacency-aware layout metrics.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence"], "snippet": "We introduce RealDocBench, a two-track benchmark built from real regulated documents."}, "familyId": "bmf_f035a1fb06db", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_realdocbench_6da01042", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.07401", "paper": "https://arxiv.org/abs/2606.07401", "pdf": "https://arxiv.org/pdf/2606.07401", "project": null, "report": "https://arxiv.org/abs/2606.07401"}, "metrics": [], "motivation": "Document parsing systems are increasingly deployed in high-stakes, regulated workflows such as mortgage underwriting, financial reporting, supply-chain logistics, and clinical records.", "name": "RealDocBench", "oneLine": "RealDocBench evaluates field-level QA and layout understanding on real regulated documents, with 1,356 field-level questions over 581 documents and 1,500 annotated page images, scored on per-field accuracy and adjacency-aware layout metrics.", "paperTitle": "RealDocBench: A Benchmark for Field-Level QA and Layout Understanding on Real-World Regulated Documents", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.07401", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-06-05", "source": {"authors": ["Ameya Joshi", "Joon Kim", "Gus Eggert", "Joseph Bajor", "Cindy Hao", "Jing Reyhan", "Kushal Byatnal", "Eli Badgio"], "categories": ["cs.CV"], "id": "2606.07401", "title": "RealDocBench: A Benchmark for Field-Level QA and Layout Understanding on Real-World Regulated Documents", "type": "arxiv", "url": "https://arxiv.org/abs/2606.07401"}, "sourceUpdatedAt": "2026-06-05T15:41:34Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "It addresses the gap in document parsing evaluation by focusing on real-world regulated documents and specific field-level needs, enabling cost-aware comparisons of commercial and open-source systems."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
