{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.03650", "kind": "paper", "url": "https://arxiv.org/abs/2607.03650"}, {"id": "gh:clinocr-bench/clinocr-bench", "kind": "repo", "url": "https://github.com/ClinOCR-Bench/ClinOCR-Bench"}], "authors": ["Enshuo Hsu", "Jin Zhou", "Kirk Roberts"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.03650", "source_url": "https://arxiv.org/abs/2607.03650"}, "categories": ["Biology & Drug Discovery", "Vision & 3D", "cs.CV"], "description": {"en": "ClinOCR-Bench is a public dataset of 384 scanned clinical documents across six artifact subsets (normal, handwriting, poor quality, rotation, tables, mixed). It evaluates OCR systems on clinical text extraction, with ground truth transcriptions and template-aware train/test splits supporting 0-shot and 1-shot evaluation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.03650", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.03650", "title": "ClinOCR-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["The University of Texas MD Anderson Cancer Center", "The University of Texas Health Science Center", "The University of Texas Health Science Center at Houston"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.03650", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.03650", "source_url": "https://openalex.org/W7167578445"}, "key": "claire-radar:2607.03650", "modality": null, "name": "ClinOCR-Bench", "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_clinocr-bench_8be5c7a7", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "1c768513f2d1e7c55f76566a103d4467f384a01efd555f0020682ecdbc7168d6", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.03650"}, "publisher": null, "released": "2026-07-04", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.03650", "source_url": "https://arxiv.org/abs/2607.03650"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-03650", "source": "claire_radar", "source_benchmark_id": "2607.03650", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Biology & Drug Discovery"], "area": "Vision & 3D", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/ClinOCR-Bench/ClinOCR-Bench", "githubScope": "benchmark_repo", "githubStars": 3, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2607.03650", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "fresh"}, "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": "7fa6bc0c3ec964067f111d5cdd8036b541d6ebdf4c991b8b6b834230ce37a648", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ClinOCR-Bench is a public dataset of 384 scanned clinical documents across six artifact subsets (normal, handwriting, poor quality, rotation, tables, mixed). It evaluates OCR systems on clinical text extraction, with ground truth transcriptions and template-aware train/test splits supporting 0-shot and 1-shot evaluation.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "Therefore, we release a publicly available, realistic-looking OCR benchmark dataset, ClinOCR-Bench, with 384 scanned images across 6 subsets: Normal, Handwriting, Poor Quality, Rotation, Tables, and Mix-artifacts."}, "familyId": "bmf_6c9fc7398ebb", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/ClinOCR-Bench/ClinOCR-Bench", "url": "https://github.com/ClinOCR-Bench/ClinOCR-Bench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/ClinOCR-Bench/ClinOCR-Bench", "url": "https://huggingface.co/papers/2607.03650"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/ClinOCR-Bench/ClinOCR-Bench", "url": "https://arxiv.org/abs/2607.03650"}], "scope": "benchmark_repo", "url": "https://github.com/ClinOCR-Bench/ClinOCR-Bench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_clinocr-bench_8be5c7a7", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Pharma & Biotech"], "links": {"code": "https://github.com/ClinOCR-Bench/ClinOCR-Bench", "data": null, "hfPaper": "https://huggingface.co/papers/2607.03650", "paper": "https://arxiv.org/abs/2607.03650", "pdf": "https://arxiv.org/pdf/2607.03650", "project": null, "report": "https://arxiv.org/abs/2607.03650"}, "metrics": [], "motivation": "Extracting textual information from scanned medical documents, such as external laboratory reports and manually filled forms, has been a major challenge in modern electronic health records (EHRs).", "name": "ClinOCR-Bench", "oneLine": "ClinOCR-Bench is a public dataset of 384 scanned clinical documents across six artifact subsets (normal, handwriting, poor quality, rotation, tables, mixed). It evaluates OCR systems on clinical text extraction, with ground truth transcriptions and template-aware train/test splits supporting 0-shot and 1-shot evaluation.", "paperTitle": "ClinOCR-Bench: A Comprehensive Clinical Scanned Document Dataset for Optical Character Recognition Model Evaluation", "primaryDomain": "Biology & Drug Discovery", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.03650", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-04", "source": {"authors": ["Enshuo Hsu", "Jin Zhou", "Kirk Roberts"], "categories": ["cs.CV", "cs.AI"], "id": "2607.03650", "title": "ClinOCR-Bench: A Comprehensive Clinical Scanned Document Dataset for Optical Character Recognition Model Evaluation", "type": "arxiv", "url": "https://arxiv.org/abs/2607.03650"}, "sourceUpdatedAt": "2026-07-04T00:19:31Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Existing clinical OCR evaluations rely on private data lacking common scan artifacts. ClinOCR-Bench provides a standardized, realistic set of clinical documents with controlled artifacts, enabling reproducible comparison of OCR and vision-language models on real-world clinical scanning challenges."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
