{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CTZASW2CD662MTFTMTLKZCP7T4","short_pith_number":"pith:CTZASW2C","schema_version":"1.0","canonical_sha256":"14f2095b421fbda64cb364d6ac89ff9f19330d6bcdc5218ccc20ad7b1f2f14a0","source":{"kind":"arxiv","id":"2607.20848","version":1},"attestation_state":"computed","paper":{"title":"Auditing Evidence Use in Medical LLM Diagnosis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fuji Ren, Jiawen Deng, Junchi Liao","submitted_at":"2026-07-23T02:20:15Z","abstract_excerpt":"Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the case evidence appropriately. We present a behavioral audit of evidence use in medical diagnosis. For each case, we decompose patient information into evidence units, score candidate diagnoses under controlled evidence subsets, and mine low-order interactions in diagnostic margins. Because medical evidence is diagnosis-relative, the audit separates interaction discovery from failure assignment: large or negative interactions can reflect plausible "},"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":"2607.20848","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T02:20:15Z","cross_cats_sorted":[],"title_canon_sha256":"4696fd5c7b6721ca9cf09f2b2fb85764cc3a9fec4c6f2eb15d234f7209c734f7","abstract_canon_sha256":"02bc7bb44fc8d8c1a2f1fdf254c7a03a39914751b5703b67a96f2e54cd181489"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T00:23:38.057038Z","signature_b64":"UBmwlLVvKs2UAdroQAyOWFL+yAoGBQm5eH0MUKj2BF+N8ZG/fe21EudSyT9iku3LbhDzf7yNEDQpEcMONPiNCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"14f2095b421fbda64cb364d6ac89ff9f19330d6bcdc5218ccc20ad7b1f2f14a0","last_reissued_at":"2026-07-24T00:23:38.056157Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T00:23:38.056157Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Auditing Evidence Use in Medical LLM Diagnosis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fuji Ren, Jiawen Deng, Junchi Liao","submitted_at":"2026-07-23T02:20:15Z","abstract_excerpt":"Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the case evidence appropriately. We present a behavioral audit of evidence use in medical diagnosis. For each case, we decompose patient information into evidence units, score candidate diagnoses under controlled evidence subsets, and mine low-order interactions in diagnostic margins. Because medical evidence is diagnosis-relative, the audit separates interaction discovery from failure assignment: large or negative interactions can reflect plausible "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20848","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/2607.20848/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":"2607.20848","created_at":"2026-07-24T00:23:38.056619+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.20848v1","created_at":"2026-07-24T00:23:38.056619+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20848","created_at":"2026-07-24T00:23:38.056619+00:00"},{"alias_kind":"pith_short_12","alias_value":"CTZASW2CD662","created_at":"2026-07-24T00:23:38.056619+00:00"},{"alias_kind":"pith_short_16","alias_value":"CTZASW2CD662MTFT","created_at":"2026-07-24T00:23:38.056619+00:00"},{"alias_kind":"pith_short_8","alias_value":"CTZASW2C","created_at":"2026-07-24T00:23:38.056619+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4","json":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4.json","graph_json":"https://pith.science/api/pith-number/CTZASW2CD662MTFTMTLKZCP7T4/graph.json","events_json":"https://pith.science/api/pith-number/CTZASW2CD662MTFTMTLKZCP7T4/events.json","paper":"https://pith.science/paper/CTZASW2C"},"agent_actions":{"view_html":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4","download_json":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4.json","view_paper":"https://pith.science/paper/CTZASW2C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.20848&json=true","fetch_graph":"https://pith.science/api/pith-number/CTZASW2CD662MTFTMTLKZCP7T4/graph.json","fetch_events":"https://pith.science/api/pith-number/CTZASW2CD662MTFTMTLKZCP7T4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4/action/storage_attestation","attest_author":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4/action/author_attestation","sign_citation":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4/action/citation_signature","submit_replication":"https://pith.science/pith/CTZASW2CD662MTFTMTLKZCP7T4/action/replication_record"}},"created_at":"2026-07-24T00:23:38.056619+00:00","updated_at":"2026-07-24T00:23:38.056619+00:00"}