{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YLF3BTXGTTAWKSTF5AHCHAZBZ7","short_pith_number":"pith:YLF3BTXG","schema_version":"1.0","canonical_sha256":"c2cbb0cee69cc1654a65e80e238321cfca4351e3d092dfb3f732c9d34cc46914","source":{"kind":"arxiv","id":"2606.21517","version":1},"attestation_state":"computed","paper":{"title":"MedHal-Loc: Are \"Explainable-by-Architecture\" Medical Hallucination Detectors Faithful Localizers? A Localization Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Daojian Lu, Fengdan Chen, Jvyu Cai, Minmin Chen, Yining Dai","submitted_at":"2026-06-19T15:11:37Z","abstract_excerpt":"Detecting hallucinations in clinical text is increasingly framed as an explainability problem: systems should not merely flag an unreliable response but point to the offending span. Architectures built around knowledge-graph (KG) triple decomposition are marketed for exactly this auditability, yet their localization ability is typically assumed rather than measured. We introduce MedHal-Loc, a benchmark and metric for localization faithfulness -- whether a detector's top-ranked error unit actually overlaps the erroneous span. The controlled subset comprises 300 PubMedQA-derived statements with "},"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":"2606.21517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-19T15:11:37Z","cross_cats_sorted":[],"title_canon_sha256":"a5858586a52cd8133f02f3e9421cc21d250528d9eda1644160c98534cf4f36fc","abstract_canon_sha256":"d7bb308f09fe2aaa52066c18a706af75580293a82aae426a2a98ba1432c9e791"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:13:13.270546Z","signature_b64":"/HJQimUVXQIHA/b+I0s2Y7m0ToIxnuP+OFJxmgZKFHibbp0nH/ktM9NkqnJyaQvmbb9bN4OdgIJdBfKiyRS5BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2cbb0cee69cc1654a65e80e238321cfca4351e3d092dfb3f732c9d34cc46914","last_reissued_at":"2026-06-23T01:13:13.269955Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:13:13.269955Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MedHal-Loc: Are \"Explainable-by-Architecture\" Medical Hallucination Detectors Faithful Localizers? A Localization Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Daojian Lu, Fengdan Chen, Jvyu Cai, Minmin Chen, Yining Dai","submitted_at":"2026-06-19T15:11:37Z","abstract_excerpt":"Detecting hallucinations in clinical text is increasingly framed as an explainability problem: systems should not merely flag an unreliable response but point to the offending span. Architectures built around knowledge-graph (KG) triple decomposition are marketed for exactly this auditability, yet their localization ability is typically assumed rather than measured. We introduce MedHal-Loc, a benchmark and metric for localization faithfulness -- whether a detector's top-ranked error unit actually overlaps the erroneous span. The controlled subset comprises 300 PubMedQA-derived statements with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21517","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/2606.21517/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":"2606.21517","created_at":"2026-06-23T01:13:13.270036+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21517v1","created_at":"2026-06-23T01:13:13.270036+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21517","created_at":"2026-06-23T01:13:13.270036+00:00"},{"alias_kind":"pith_short_12","alias_value":"YLF3BTXGTTAW","created_at":"2026-06-23T01:13:13.270036+00:00"},{"alias_kind":"pith_short_16","alias_value":"YLF3BTXGTTAWKSTF","created_at":"2026-06-23T01:13:13.270036+00:00"},{"alias_kind":"pith_short_8","alias_value":"YLF3BTXG","created_at":"2026-06-23T01:13:13.270036+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/YLF3BTXGTTAWKSTF5AHCHAZBZ7","json":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7.json","graph_json":"https://pith.science/api/pith-number/YLF3BTXGTTAWKSTF5AHCHAZBZ7/graph.json","events_json":"https://pith.science/api/pith-number/YLF3BTXGTTAWKSTF5AHCHAZBZ7/events.json","paper":"https://pith.science/paper/YLF3BTXG"},"agent_actions":{"view_html":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7","download_json":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7.json","view_paper":"https://pith.science/paper/YLF3BTXG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21517&json=true","fetch_graph":"https://pith.science/api/pith-number/YLF3BTXGTTAWKSTF5AHCHAZBZ7/graph.json","fetch_events":"https://pith.science/api/pith-number/YLF3BTXGTTAWKSTF5AHCHAZBZ7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7/action/storage_attestation","attest_author":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7/action/author_attestation","sign_citation":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7/action/citation_signature","submit_replication":"https://pith.science/pith/YLF3BTXGTTAWKSTF5AHCHAZBZ7/action/replication_record"}},"created_at":"2026-06-23T01:13:13.270036+00:00","updated_at":"2026-06-23T01:13:13.270036+00:00"}