{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WU3XJCXHWFRE4ZTOLZUZHAS2I6","short_pith_number":"pith:WU3XJCXH","schema_version":"1.0","canonical_sha256":"b537748ae7b1624e666e5e6993825a478f9a333f44f76d58c86036591d6096f5","source":{"kind":"arxiv","id":"2403.01002","version":2},"attestation_state":"computed","paper":{"title":"Attribute Structuring Improves LLM-Based Evaluation of Clinical Text Summaries","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chandan Singh, Hoifung Poon, Jianfeng Gao, Paul Vozila, Praveen Subramanian, Sheng Zhang, Tristan Naumann, Yiqing Xie, Zelalem Gero","submitted_at":"2024-03-01T21:59:03Z","abstract_excerpt":"Summarizing clinical text is crucial in health decision-support and clinical research. Large language models (LLMs) have shown the potential to generate accurate clinical text summaries, but still struggle with issues regarding grounding and evaluation, especially in safety-critical domains such as health. Holistically evaluating text summaries is challenging because they may contain unsubstantiated information. Here, we explore a general mitigation framework using Attribute Structuring (AS), which structures the summary evaluation process. It decomposes the evaluation process into a grounded "},"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":"2403.01002","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-01T21:59:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"223358854aa33947aed11964d286a7cc0ad56f70435a2abaf8eb30266ef437ef","abstract_canon_sha256":"856829c0e675e585cf45f383eb5d873c91f4a20bd7201b26936f82b18223e504"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:02.756728Z","signature_b64":"2sMYdYTu5nZ3h/FNqXNZNjvp2w021TyQiTNfCJ3EgW03V3cYILTJuDJfu0O3QyriZZ1K3xsBWtnQa7gTIYTeDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b537748ae7b1624e666e5e6993825a478f9a333f44f76d58c86036591d6096f5","last_reissued_at":"2026-07-05T09:49:02.756192Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:02.756192Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Attribute Structuring Improves LLM-Based Evaluation of Clinical Text Summaries","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chandan Singh, Hoifung Poon, Jianfeng Gao, Paul Vozila, Praveen Subramanian, Sheng Zhang, Tristan Naumann, Yiqing Xie, Zelalem Gero","submitted_at":"2024-03-01T21:59:03Z","abstract_excerpt":"Summarizing clinical text is crucial in health decision-support and clinical research. Large language models (LLMs) have shown the potential to generate accurate clinical text summaries, but still struggle with issues regarding grounding and evaluation, especially in safety-critical domains such as health. Holistically evaluating text summaries is challenging because they may contain unsubstantiated information. Here, we explore a general mitigation framework using Attribute Structuring (AS), which structures the summary evaluation process. It decomposes the evaluation process into a grounded "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.01002","kind":"arxiv","version":2},"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/2403.01002/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":"2403.01002","created_at":"2026-07-05T09:49:02.756260+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.01002v2","created_at":"2026-07-05T09:49:02.756260+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.01002","created_at":"2026-07-05T09:49:02.756260+00:00"},{"alias_kind":"pith_short_12","alias_value":"WU3XJCXHWFRE","created_at":"2026-07-05T09:49:02.756260+00:00"},{"alias_kind":"pith_short_16","alias_value":"WU3XJCXHWFRE4ZTO","created_at":"2026-07-05T09:49:02.756260+00:00"},{"alias_kind":"pith_short_8","alias_value":"WU3XJCXH","created_at":"2026-07-05T09:49:02.756260+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08969","citing_title":"CARE: A Conformal Safety Layer for Medical Summarization","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6","json":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6.json","graph_json":"https://pith.science/api/pith-number/WU3XJCXHWFRE4ZTOLZUZHAS2I6/graph.json","events_json":"https://pith.science/api/pith-number/WU3XJCXHWFRE4ZTOLZUZHAS2I6/events.json","paper":"https://pith.science/paper/WU3XJCXH"},"agent_actions":{"view_html":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6","download_json":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6.json","view_paper":"https://pith.science/paper/WU3XJCXH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.01002&json=true","fetch_graph":"https://pith.science/api/pith-number/WU3XJCXHWFRE4ZTOLZUZHAS2I6/graph.json","fetch_events":"https://pith.science/api/pith-number/WU3XJCXHWFRE4ZTOLZUZHAS2I6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6/action/storage_attestation","attest_author":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6/action/author_attestation","sign_citation":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6/action/citation_signature","submit_replication":"https://pith.science/pith/WU3XJCXHWFRE4ZTOLZUZHAS2I6/action/replication_record"}},"created_at":"2026-07-05T09:49:02.756260+00:00","updated_at":"2026-07-05T09:49:02.756260+00:00"}