{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QXFZ6SCTECUJSEDZRZG74MV72K","short_pith_number":"pith:QXFZ6SCT","schema_version":"1.0","canonical_sha256":"85cb9f485320a89910798e4dfe32bfd2968519a47ae5377b06679e59c7c2c154","source":{"kind":"arxiv","id":"2506.11129","version":1},"attestation_state":"computed","paper":{"title":"Trustworthy AI for Medicine: Continuous Hallucination Detection and Elimination with CHECK","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Aakash Tripathi, Carlos Garcia-Fernandez, Ghulam Rasool, Gilmer Valdes, Issam El Naqa, Luis Felipe, Monique Shotande, Muntasir Zitu, Vivek Rudrapatna","submitted_at":"2025-06-10T17:12:28Z","abstract_excerpt":"Large language models (LLMs) show promise in healthcare, but hallucinations remain a major barrier to clinical use. We present CHECK, a continuous-learning framework that integrates structured clinical databases with a classifier grounded in information theory to detect both factual and reasoning-based hallucinations. Evaluated on 1500 questions from 100 pivotal clinical trials, CHECK reduced LLama3.3-70B-Instruct hallucination rates from 31% to 0.3% - making an open source model state of the art. Its classifier generalized across medical benchmarks, achieving AUCs of 0.95-0.96, including on t"},"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":"2506.11129","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-10T17:12:28Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f3cc6e25b059e7cbcca41693452e7772c30a7842bca3212572b09f5c58163a7b","abstract_canon_sha256":"438e190d0bdd2a0862ac6a149415f459dcf4f20c09bc847ca04d9f8bbb922d5f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:00.067292Z","signature_b64":"Rf9CKUF0ZggF9XhNXJwpuKCwT+CIGFVmq/6fmkwhzCn8mz6UKV0oYjt7VjUeezTRwAqlleqLA4NIXcNjP939Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85cb9f485320a89910798e4dfe32bfd2968519a47ae5377b06679e59c7c2c154","last_reissued_at":"2026-07-05T11:21:00.066738Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:00.066738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trustworthy AI for Medicine: Continuous Hallucination Detection and Elimination with CHECK","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Aakash Tripathi, Carlos Garcia-Fernandez, Ghulam Rasool, Gilmer Valdes, Issam El Naqa, Luis Felipe, Monique Shotande, Muntasir Zitu, Vivek Rudrapatna","submitted_at":"2025-06-10T17:12:28Z","abstract_excerpt":"Large language models (LLMs) show promise in healthcare, but hallucinations remain a major barrier to clinical use. We present CHECK, a continuous-learning framework that integrates structured clinical databases with a classifier grounded in information theory to detect both factual and reasoning-based hallucinations. Evaluated on 1500 questions from 100 pivotal clinical trials, CHECK reduced LLama3.3-70B-Instruct hallucination rates from 31% to 0.3% - making an open source model state of the art. Its classifier generalized across medical benchmarks, achieving AUCs of 0.95-0.96, including on t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11129","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/2506.11129/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":"2506.11129","created_at":"2026-07-05T11:21:00.066815+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11129v1","created_at":"2026-07-05T11:21:00.066815+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11129","created_at":"2026-07-05T11:21:00.066815+00:00"},{"alias_kind":"pith_short_12","alias_value":"QXFZ6SCTECUJ","created_at":"2026-07-05T11:21:00.066815+00:00"},{"alias_kind":"pith_short_16","alias_value":"QXFZ6SCTECUJSEDZ","created_at":"2026-07-05T11:21:00.066815+00:00"},{"alias_kind":"pith_short_8","alias_value":"QXFZ6SCT","created_at":"2026-07-05T11:21:00.066815+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.20869","citing_title":"Clinical Reasoning AI for Oncology Treatment Planning: A Multi-Specialty Case-Based Evaluation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06028","citing_title":"A Multi-Stage Validation Framework for Trustworthy Large-scale Clinical Information Extraction using Large Language Models","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K","json":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K.json","graph_json":"https://pith.science/api/pith-number/QXFZ6SCTECUJSEDZRZG74MV72K/graph.json","events_json":"https://pith.science/api/pith-number/QXFZ6SCTECUJSEDZRZG74MV72K/events.json","paper":"https://pith.science/paper/QXFZ6SCT"},"agent_actions":{"view_html":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K","download_json":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K.json","view_paper":"https://pith.science/paper/QXFZ6SCT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11129&json=true","fetch_graph":"https://pith.science/api/pith-number/QXFZ6SCTECUJSEDZRZG74MV72K/graph.json","fetch_events":"https://pith.science/api/pith-number/QXFZ6SCTECUJSEDZRZG74MV72K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K/action/storage_attestation","attest_author":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K/action/author_attestation","sign_citation":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K/action/citation_signature","submit_replication":"https://pith.science/pith/QXFZ6SCTECUJSEDZRZG74MV72K/action/replication_record"}},"created_at":"2026-07-05T11:21:00.066815+00:00","updated_at":"2026-07-05T11:21:00.066815+00:00"}