{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WJIXVSPOZO2TUBF4WPY4DZTDTW","short_pith_number":"pith:WJIXVSPO","schema_version":"1.0","canonical_sha256":"b2517ac9eecbb53a04bcb3f1c1e6639dbabf4b078dfde3eb32e2e042c0af4e0e","source":{"kind":"arxiv","id":"2502.14302","version":1},"attestation_state":"computed","paper":{"title":"MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jiawei Xu, Junyuan Hong, Kaidi Xu, Shrey Pandit, Tianlong Chen, Ying Ding, Zhangyang Wang","submitted_at":"2025-02-20T06:33:23Z","abstract_excerpt":"Advancements in Large Language Models (LLMs) and their increasing use in medical question-answering necessitate rigorous evaluation of their reliability. A critical challenge lies in hallucination, where models generate plausible yet factually incorrect outputs. In the medical domain, this poses serious risks to patient safety and clinical decision-making. To address this, we introduce MedHallu, the first benchmark specifically designed for medical hallucination detection. MedHallu comprises 10,000 high-quality question-answer pairs derived from PubMedQA, with hallucinated answers systematical"},"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":"2502.14302","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T06:33:23Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"23873cc5e0774a99b2af85a0560055a5631cdb01540c3524efc8239cf55890b5","abstract_canon_sha256":"688e97685641bf3a79598e2258c608221d276acde5a00f24fecc990d974bd7e8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:17:24.149362Z","signature_b64":"516ZoY2Xcbr7RTtmYrQEvC2OAM+rX9lYS1UtOdkRf4vD28bvcBao+xinWEKMJJguW1ByoSvODhl+ol7dOSBZAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2517ac9eecbb53a04bcb3f1c1e6639dbabf4b078dfde3eb32e2e042c0af4e0e","last_reissued_at":"2026-07-05T10:17:24.148766Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:17:24.148766Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jiawei Xu, Junyuan Hong, Kaidi Xu, Shrey Pandit, Tianlong Chen, Ying Ding, Zhangyang Wang","submitted_at":"2025-02-20T06:33:23Z","abstract_excerpt":"Advancements in Large Language Models (LLMs) and their increasing use in medical question-answering necessitate rigorous evaluation of their reliability. A critical challenge lies in hallucination, where models generate plausible yet factually incorrect outputs. In the medical domain, this poses serious risks to patient safety and clinical decision-making. To address this, we introduce MedHallu, the first benchmark specifically designed for medical hallucination detection. MedHallu comprises 10,000 high-quality question-answer pairs derived from PubMedQA, with hallucinated answers systematical"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14302","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/2502.14302/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":"2502.14302","created_at":"2026-07-05T10:17:24.148838+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.14302v1","created_at":"2026-07-05T10:17:24.148838+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14302","created_at":"2026-07-05T10:17:24.148838+00:00"},{"alias_kind":"pith_short_12","alias_value":"WJIXVSPOZO2T","created_at":"2026-07-05T10:17:24.148838+00:00"},{"alias_kind":"pith_short_16","alias_value":"WJIXVSPOZO2TUBF4","created_at":"2026-07-05T10:17:24.148838+00:00"},{"alias_kind":"pith_short_8","alias_value":"WJIXVSPO","created_at":"2026-07-05T10:17:24.148838+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01301","citing_title":"Med-HEAL: Analyzing and Mitigating Hallucinations in Medical LLMs with Hallucination-Aware In-Context Learning","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22963","citing_title":"Graph Alignment Topology as an Inductive Bias for Grounding Detection","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04180","citing_title":"MedFabric and EtHER: A Data-Centric Framework for Word-Level Fabrication Generation and Detection in Medical LLMs","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07209","citing_title":"Hallucination Detection via Activations of Open-Weight Proxy Analyzers","ref_index":17,"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":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW","json":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW.json","graph_json":"https://pith.science/api/pith-number/WJIXVSPOZO2TUBF4WPY4DZTDTW/graph.json","events_json":"https://pith.science/api/pith-number/WJIXVSPOZO2TUBF4WPY4DZTDTW/events.json","paper":"https://pith.science/paper/WJIXVSPO"},"agent_actions":{"view_html":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW","download_json":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW.json","view_paper":"https://pith.science/paper/WJIXVSPO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.14302&json=true","fetch_graph":"https://pith.science/api/pith-number/WJIXVSPOZO2TUBF4WPY4DZTDTW/graph.json","fetch_events":"https://pith.science/api/pith-number/WJIXVSPOZO2TUBF4WPY4DZTDTW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW/action/storage_attestation","attest_author":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW/action/author_attestation","sign_citation":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW/action/citation_signature","submit_replication":"https://pith.science/pith/WJIXVSPOZO2TUBF4WPY4DZTDTW/action/replication_record"}},"created_at":"2026-07-05T10:17:24.148838+00:00","updated_at":"2026-07-05T10:17:24.148838+00:00"}