{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.14697", "kind": "paper", "url": "https://arxiv.org/abs/2606.14697"}, {"id": "gh:alibaba-damo-academy/clinhallu", "kind": "repo", "url": "https://github.com/alibaba-damo-academy/ClinHallu"}], "authors": ["Sicheng Yang", "Hangjie Yuan", "Wenjun Zhang", "Jinwang Wang", "Yichen Qian", "Weihua Chen", "Fan Wang", "Lei Zhu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.14697", "source_url": "https://arxiv.org/abs/2606.14697"}, "categories": ["Biology & Drug Discovery", "Reasoning", "Vision & 3D"], "description": {"en": "ClinHallu is a benchmark for diagnosing stage-wise hallucinations in medical multimodal large language models. It contains 7,031 instances with structured reasoning traces decomposed into visual recognition, knowledge recall, and reasoning integration, along with stage-replacement interventions for measuring final answer changes."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.14697", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.14697", "title": "ClinHallu"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.14697", "modality": null, "name": "ClinHallu", "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_clinhallu_85f948a0", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "c5c9bf9595262da34f8cccdc542cb583f8c7c01fc8492d91c266e24226508784", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.14697"}, "publisher": null, "released": "2026-06-12", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.14697", "source_url": "https://arxiv.org/abs/2606.14697"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-14697", "source": "claire_radar", "source_benchmark_id": "2606.14697", "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/alibaba-damo-academy/ClinHallu", "githubScope": "benchmark_repo", "githubStars": 10, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 7, "hfPaperUrl": "https://huggingface.co/papers/2606.14697", "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": "2026-10-01", "state": "stale"}}, "status": "fresh"}, "capabilities": ["Reasoning"], "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": "c8e07fb367894c41ba360c6edec112e26ee266686dcf9cb604385753e9d977d2", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ClinHallu is a benchmark for diagnosing stage-wise hallucinations in medical multimodal large language models. It contains 7,031 instances with structured reasoning traces decomposed into visual recognition, knowledge recall, and reasoning integration, along with stage-replacement interventions for measuring final answer changes.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To enable source-level hallucination diagnosis, we introduce ClinHallu, a benchmark for stage-wise hallucination diagnosis in medical MLLM reasoning."}, "familyId": "bmf_52fa2852cf5f", "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/alibaba-damo-academy/ClinHallu", "url": "https://github.com/alibaba-damo-academy/ClinHallu"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/alibaba-damo-academy/ClinHallu", "url": "https://huggingface.co/papers/2606.14697"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/alibaba-damo-academy/ClinHallu", "url": "https://huggingface.co/api/papers/2606.14697"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/alibaba-damo-academy/ClinHallu.", "url": "https://arxiv.org/abs/2606.14697"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/alibaba-damo-academy/ClinHallu", "url": "https://arxiv.org/abs/2606.14697"}], "scope": "benchmark_repo", "url": "https://github.com/alibaba-damo-academy/ClinHallu", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_clinhallu_85f948a0", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Pharma & Biotech"], "links": {"code": "https://github.com/alibaba-damo-academy/ClinHallu", "data": null, "hfPaper": "https://huggingface.co/papers/2606.14697", "paper": "https://arxiv.org/abs/2606.14697", "pdf": "https://arxiv.org/pdf/2606.14697", "project": null, "report": "https://arxiv.org/abs/2606.14697"}, "metrics": [], "motivation": "Building trustworthy medical multimodal large language models (MLLMs) is critical for reliable clinical decision support.", "name": "ClinHallu", "oneLine": "ClinHallu is a benchmark for diagnosing stage-wise hallucinations in medical multimodal large language models. It contains 7,031 instances with structured reasoning traces decomposed into visual recognition, knowledge recall, and reasoning integration, along with stage-replacement interventions for measuring final answer changes.", "paperTitle": "ClinHallu: A Benchmark for Diagnosing Stage-Wise Hallucinations in Medical MLLM Reasoning", "primaryDomain": "Biology & Drug Discovery", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.14697", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Alibaba DAMO Academy", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/alibaba-damo-academy/ClinHallu"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-12", "source": {"authors": ["Sicheng Yang", "Hangjie Yuan", "Wenjun Zhang", "Jinwang Wang", "Yichen Qian", "Weihua Chen", "Fan Wang", "Lei Zhu"], "categories": ["cs.CV", "cs.AI", "cs.CL"], "id": "2606.14697", "title": "ClinHallu: A Benchmark for Diagnosing Stage-Wise Hallucinations in Medical MLLM Reasoning", "type": "arxiv", "url": "https://arxiv.org/abs/2606.14697"}, "sourceUpdatedAt": "2026-06-12T17:58:38Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Existing medical hallucination benchmarks often ignore the source of hallucinations within reasoning. ClinHallu allows for fine-grained diagnosis of where errors originate, providing a testbed for improving model reliability in clinical decision support."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
