{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EBABOF74UFSY4IA3UQWN6AZKRJ","short_pith_number":"pith:EBABOF74","schema_version":"1.0","canonical_sha256":"20401717fca1658e201ba42cdf032a8a60fa6b326e070a0f5e1c783aa095ebac","source":{"kind":"arxiv","id":"2509.10843","version":1},"attestation_state":"computed","paper":{"title":"Evaluating Large Language Models for Evidence-Based Clinical Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Can Wang, Yiqun Chen","submitted_at":"2025-09-13T15:03:34Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated substantial progress in biomedical and clinical applications, motivating rigorous evaluation of their ability to answer nuanced, evidence-based questions. We curate a multi-source benchmark drawing from Cochrane systematic reviews and clinical guidelines, including structured recommendations from the American Heart Association and narrative guidance used by insurers. Using GPT-4o-mini and GPT-5, we observe consistent performance patterns across sources and clinical domains: accuracy is highest on structured guideline recommendations (90%) and lowe"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2509.10843","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-13T15:03:34Z","cross_cats_sorted":[],"title_canon_sha256":"2542b988ecd2d2adbd9530844602b56df66a5d8bc61ed8e6fc3c7f9c100fdc98","abstract_canon_sha256":"15395aeed66e3cf77a448acf52ad7c9e810f0120b09539972d1b80ee72dcc769"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:29.026811Z","signature_b64":"Q2Dlg83H9A8CcsolJS1xMeChYTJaxu1T35NDHc2toD/X5ZYbbmBvUvPhl59N5N+X2TYmGqjxCNh+W/MmGiQ0Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20401717fca1658e201ba42cdf032a8a60fa6b326e070a0f5e1c783aa095ebac","last_reissued_at":"2026-07-05T12:11:29.026245Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:29.026245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating Large Language Models for Evidence-Based Clinical Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Can Wang, Yiqun Chen","submitted_at":"2025-09-13T15:03:34Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated substantial progress in biomedical and clinical applications, motivating rigorous evaluation of their ability to answer nuanced, evidence-based questions. We curate a multi-source benchmark drawing from Cochrane systematic reviews and clinical guidelines, including structured recommendations from the American Heart Association and narrative guidance used by insurers. Using GPT-4o-mini and GPT-5, we observe consistent performance patterns across sources and clinical domains: accuracy is highest on structured guideline recommendations (90%) and lowe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.10843","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.10843/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2509.10843","created_at":"2026-07-05T12:11:29.026301+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.10843v1","created_at":"2026-07-05T12:11:29.026301+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.10843","created_at":"2026-07-05T12:11:29.026301+00:00"},{"alias_kind":"pith_short_12","alias_value":"EBABOF74UFSY","created_at":"2026-07-05T12:11:29.026301+00:00"},{"alias_kind":"pith_short_16","alias_value":"EBABOF74UFSY4IA3","created_at":"2026-07-05T12:11:29.026301+00:00"},{"alias_kind":"pith_short_8","alias_value":"EBABOF74","created_at":"2026-07-05T12:11:29.026301+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.06668","citing_title":"Contradictions in Context: Challenges for Retrieval-Augmented Generation in Healthcare","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ","json":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ.json","graph_json":"https://pith.science/api/pith-number/EBABOF74UFSY4IA3UQWN6AZKRJ/graph.json","events_json":"https://pith.science/api/pith-number/EBABOF74UFSY4IA3UQWN6AZKRJ/events.json","paper":"https://pith.science/paper/EBABOF74"},"agent_actions":{"view_html":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ","download_json":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ.json","view_paper":"https://pith.science/paper/EBABOF74","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.10843&json=true","fetch_graph":"https://pith.science/api/pith-number/EBABOF74UFSY4IA3UQWN6AZKRJ/graph.json","fetch_events":"https://pith.science/api/pith-number/EBABOF74UFSY4IA3UQWN6AZKRJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ/action/storage_attestation","attest_author":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ/action/author_attestation","sign_citation":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ/action/citation_signature","submit_replication":"https://pith.science/pith/EBABOF74UFSY4IA3UQWN6AZKRJ/action/replication_record"}},"created_at":"2026-07-05T12:11:29.026301+00:00","updated_at":"2026-07-05T12:11:29.026301+00:00"}