{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MXU4TJDVDXUJ4VVZPWVVIWOFKG","short_pith_number":"pith:MXU4TJDV","schema_version":"1.0","canonical_sha256":"65e9c9a4751de89e56b97dab5459c551b9c8a28bdcada4d586e72bf593c3cb6d","source":{"kind":"arxiv","id":"2506.18387","version":1},"attestation_state":"computed","paper":{"title":"Evaluating Causal Explanation in Medical Reports with LLM-Based and Human-Aligned Metrics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Key-Sun Choi, Yousang Cho","submitted_at":"2025-06-23T08:19:21Z","abstract_excerpt":"This study investigates how accurately different evaluation metrics capture the quality of causal explanations in automatically generated diagnostic reports. We compare six metrics: BERTScore, Cosine Similarity, BioSentVec, GPT-White, GPT-Black, and expert qualitative assessment across two input types: observation-based and multiple-choice-based report generation. Two weighting strategies are applied: one reflecting task-specific priorities, and the other assigning equal weights to all metrics. Our results show that GPT-Black demonstrates the strongest discriminative power in identifying logic"},"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":"2506.18387","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T08:19:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1afd3698d80939840986ba0d2da09ad274577312195338a9164742e50c97bfff","abstract_canon_sha256":"bff821ebe7c91c46aaeb3db3c738772db43ef08b4b620c59dd52d42a23279617"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:49.592324Z","signature_b64":"infS6UrkT2Z4GnNWEYLJ+/pd3xTfYECSImEUqaAl2fWoZlZmOvRhEq2n9qnD/HeL7SFbDpg3pD27PKe8OGleCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"65e9c9a4751de89e56b97dab5459c551b9c8a28bdcada4d586e72bf593c3cb6d","last_reissued_at":"2026-07-05T11:25:49.591896Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:49.591896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating Causal Explanation in Medical Reports with LLM-Based and Human-Aligned Metrics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Key-Sun Choi, Yousang Cho","submitted_at":"2025-06-23T08:19:21Z","abstract_excerpt":"This study investigates how accurately different evaluation metrics capture the quality of causal explanations in automatically generated diagnostic reports. We compare six metrics: BERTScore, Cosine Similarity, BioSentVec, GPT-White, GPT-Black, and expert qualitative assessment across two input types: observation-based and multiple-choice-based report generation. Two weighting strategies are applied: one reflecting task-specific priorities, and the other assigning equal weights to all metrics. Our results show that GPT-Black demonstrates the strongest discriminative power in identifying logic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18387","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/2506.18387/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":"2506.18387","created_at":"2026-07-05T11:25:49.591951+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.18387v1","created_at":"2026-07-05T11:25:49.591951+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18387","created_at":"2026-07-05T11:25:49.591951+00:00"},{"alias_kind":"pith_short_12","alias_value":"MXU4TJDVDXUJ","created_at":"2026-07-05T11:25:49.591951+00:00"},{"alias_kind":"pith_short_16","alias_value":"MXU4TJDVDXUJ4VVZ","created_at":"2026-07-05T11:25:49.591951+00:00"},{"alias_kind":"pith_short_8","alias_value":"MXU4TJDV","created_at":"2026-07-05T11:25:49.591951+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/MXU4TJDVDXUJ4VVZPWVVIWOFKG","json":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG.json","graph_json":"https://pith.science/api/pith-number/MXU4TJDVDXUJ4VVZPWVVIWOFKG/graph.json","events_json":"https://pith.science/api/pith-number/MXU4TJDVDXUJ4VVZPWVVIWOFKG/events.json","paper":"https://pith.science/paper/MXU4TJDV"},"agent_actions":{"view_html":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG","download_json":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG.json","view_paper":"https://pith.science/paper/MXU4TJDV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.18387&json=true","fetch_graph":"https://pith.science/api/pith-number/MXU4TJDVDXUJ4VVZPWVVIWOFKG/graph.json","fetch_events":"https://pith.science/api/pith-number/MXU4TJDVDXUJ4VVZPWVVIWOFKG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG/action/storage_attestation","attest_author":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG/action/author_attestation","sign_citation":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG/action/citation_signature","submit_replication":"https://pith.science/pith/MXU4TJDVDXUJ4VVZPWVVIWOFKG/action/replication_record"}},"created_at":"2026-07-05T11:25:49.591951+00:00","updated_at":"2026-07-05T11:25:49.591951+00:00"}