{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RJLWBRIRWLHPZFAC6JDHZAKMRD","short_pith_number":"pith:RJLWBRIR","schema_version":"1.0","canonical_sha256":"8a5760c511b2cefc9402f2467c814c88dfd1d50408944d340c83b80f4ee674a6","source":{"kind":"arxiv","id":"2508.19096","version":1},"attestation_state":"computed","paper":{"title":"Trustworthy Agents for Electronic Health Records through Confidence Estimation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Minbyul Jeong, Mujeen Sung, Yongwoo Song","submitted_at":"2025-08-26T14:59:04Z","abstract_excerpt":"Large language models (LLMs) show promise for extracting information from Electronic Health Records (EHR) and supporting clinical decisions. However, deployment in clinical settings faces challenges due to hallucination risks. We propose Hallucination Controlled Accuracy at k% (HCAcc@k%), a novel metric quantifying the accuracy-reliability trade-off at varying confidence thresholds. We introduce TrustEHRAgent, a confidence-aware agent incorporating stepwise confidence estimation for clinical question answering. Experiments on MIMIC-III and eICU datasets show TrustEHRAgent outperforms baselines"},"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":"2508.19096","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-26T14:59:04Z","cross_cats_sorted":[],"title_canon_sha256":"b42a5ba57185c4ab48664cce8fe0a94222702943b2525eae66336e3473ef7268","abstract_canon_sha256":"4b7ebc61ffc0c0f5bc6b423b3bfaff5fe244d932f65a360e30d11c701ff3e501"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:36.635905Z","signature_b64":"2D3Zuwe+7S3nD7a4SdapeBZSwxfLFqRkO5C5qRL+RtqWbaclcAZK/gedheKANW7hygHf0WFS1X4GsmyA0RiLCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a5760c511b2cefc9402f2467c814c88dfd1d50408944d340c83b80f4ee674a6","last_reissued_at":"2026-07-05T11:59:36.635446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:36.635446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trustworthy Agents for Electronic Health Records through Confidence Estimation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Minbyul Jeong, Mujeen Sung, Yongwoo Song","submitted_at":"2025-08-26T14:59:04Z","abstract_excerpt":"Large language models (LLMs) show promise for extracting information from Electronic Health Records (EHR) and supporting clinical decisions. However, deployment in clinical settings faces challenges due to hallucination risks. We propose Hallucination Controlled Accuracy at k% (HCAcc@k%), a novel metric quantifying the accuracy-reliability trade-off at varying confidence thresholds. We introduce TrustEHRAgent, a confidence-aware agent incorporating stepwise confidence estimation for clinical question answering. Experiments on MIMIC-III and eICU datasets show TrustEHRAgent outperforms baselines"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19096","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/2508.19096/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":"2508.19096","created_at":"2026-07-05T11:59:36.635505+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.19096v1","created_at":"2026-07-05T11:59:36.635505+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19096","created_at":"2026-07-05T11:59:36.635505+00:00"},{"alias_kind":"pith_short_12","alias_value":"RJLWBRIRWLHP","created_at":"2026-07-05T11:59:36.635505+00:00"},{"alias_kind":"pith_short_16","alias_value":"RJLWBRIRWLHPZFAC","created_at":"2026-07-05T11:59:36.635505+00:00"},{"alias_kind":"pith_short_8","alias_value":"RJLWBRIR","created_at":"2026-07-05T11:59:36.635505+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/RJLWBRIRWLHPZFAC6JDHZAKMRD","json":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD.json","graph_json":"https://pith.science/api/pith-number/RJLWBRIRWLHPZFAC6JDHZAKMRD/graph.json","events_json":"https://pith.science/api/pith-number/RJLWBRIRWLHPZFAC6JDHZAKMRD/events.json","paper":"https://pith.science/paper/RJLWBRIR"},"agent_actions":{"view_html":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD","download_json":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD.json","view_paper":"https://pith.science/paper/RJLWBRIR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.19096&json=true","fetch_graph":"https://pith.science/api/pith-number/RJLWBRIRWLHPZFAC6JDHZAKMRD/graph.json","fetch_events":"https://pith.science/api/pith-number/RJLWBRIRWLHPZFAC6JDHZAKMRD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD/action/storage_attestation","attest_author":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD/action/author_attestation","sign_citation":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD/action/citation_signature","submit_replication":"https://pith.science/pith/RJLWBRIRWLHPZFAC6JDHZAKMRD/action/replication_record"}},"created_at":"2026-07-05T11:59:36.635505+00:00","updated_at":"2026-07-05T11:59:36.635505+00:00"}