{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:UOBOWZJYL5AP6DBOR6BCL77IOP","short_pith_number":"pith:UOBOWZJY","schema_version":"1.0","canonical_sha256":"a382eb65385f40ff0c2e8f8225ffe873d02f4be52bb55a225bf565b60aba39fd","source":{"kind":"arxiv","id":"2401.14490","version":1},"attestation_state":"computed","paper":{"title":"LongHealth: A Question Answering Benchmark with Long Clinical Documents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexander L\\\"oser, Daniel Truhn, Felix Busch, Hugo JWL. Aerts, Jakob Nikolas Kather, Jean-Baptiste Excoffier, Keno Bressem, Lisa Adams, Matthieu Ortala, Tianyu Han","submitted_at":"2024-01-25T19:57:00Z","abstract_excerpt":"Background: Recent advancements in large language models (LLMs) offer potential benefits in healthcare, particularly in processing extensive patient records. However, existing benchmarks do not fully assess LLMs' capability in handling real-world, lengthy clinical data.\n  Methods: We present the LongHealth benchmark, comprising 20 detailed fictional patient cases across various diseases, with each case containing 5,090 to 6,754 words. The benchmark challenges LLMs with 400 multiple-choice questions in three categories: information extraction, negation, and sorting, challenging LLMs to extract "},"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":"2401.14490","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T19:57:00Z","cross_cats_sorted":[],"title_canon_sha256":"7f04d7170b4657d4bddf164ded75aa8c02be0d6289e1776dbe1f0eb5392a3efe","abstract_canon_sha256":"241d053112ac58a11df97ba1dc3ac1a13d503ef0b84fcb732724a889fa721a13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:37:57.530367Z","signature_b64":"/kUzVGmiCmiMINDv5IG8ckmEcl03JTsHFnJWEwiOiEhaVwh+WmmeSJoiwEcpNS7q2N5zq+CtlfPppctPw44eCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a382eb65385f40ff0c2e8f8225ffe873d02f4be52bb55a225bf565b60aba39fd","last_reissued_at":"2026-07-05T07:37:57.529886Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:37:57.529886Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LongHealth: A Question Answering Benchmark with Long Clinical Documents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexander L\\\"oser, Daniel Truhn, Felix Busch, Hugo JWL. Aerts, Jakob Nikolas Kather, Jean-Baptiste Excoffier, Keno Bressem, Lisa Adams, Matthieu Ortala, Tianyu Han","submitted_at":"2024-01-25T19:57:00Z","abstract_excerpt":"Background: Recent advancements in large language models (LLMs) offer potential benefits in healthcare, particularly in processing extensive patient records. However, existing benchmarks do not fully assess LLMs' capability in handling real-world, lengthy clinical data.\n  Methods: We present the LongHealth benchmark, comprising 20 detailed fictional patient cases across various diseases, with each case containing 5,090 to 6,754 words. The benchmark challenges LLMs with 400 multiple-choice questions in three categories: information extraction, negation, and sorting, challenging LLMs to extract "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14490","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/2401.14490/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":"2401.14490","created_at":"2026-07-05T07:37:57.529939+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.14490v1","created_at":"2026-07-05T07:37:57.529939+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14490","created_at":"2026-07-05T07:37:57.529939+00:00"},{"alias_kind":"pith_short_12","alias_value":"UOBOWZJYL5AP","created_at":"2026-07-05T07:37:57.529939+00:00"},{"alias_kind":"pith_short_16","alias_value":"UOBOWZJYL5AP6DBO","created_at":"2026-07-05T07:37:57.529939+00:00"},{"alias_kind":"pith_short_8","alias_value":"UOBOWZJY","created_at":"2026-07-05T07:37:57.529939+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24734","citing_title":"Task Decomposition for Efficient Annotation","ref_index":131,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05698","citing_title":"Rethinking LoRA Memory Through the Lens of KV Cache Compression","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05661","citing_title":"Continual Learning Bench: Evaluating Frontier AI Systems in Real-World Stateful Environments","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.32002","citing_title":"Self-Study Reconsidered: The Hidden Fragility of Learning from Self-Generated QA","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09778","citing_title":"Nectar: Neural Estimation of Cached-Token Attention via Regression","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP","json":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP.json","graph_json":"https://pith.science/api/pith-number/UOBOWZJYL5AP6DBOR6BCL77IOP/graph.json","events_json":"https://pith.science/api/pith-number/UOBOWZJYL5AP6DBOR6BCL77IOP/events.json","paper":"https://pith.science/paper/UOBOWZJY"},"agent_actions":{"view_html":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP","download_json":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP.json","view_paper":"https://pith.science/paper/UOBOWZJY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.14490&json=true","fetch_graph":"https://pith.science/api/pith-number/UOBOWZJYL5AP6DBOR6BCL77IOP/graph.json","fetch_events":"https://pith.science/api/pith-number/UOBOWZJYL5AP6DBOR6BCL77IOP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP/action/storage_attestation","attest_author":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP/action/author_attestation","sign_citation":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP/action/citation_signature","submit_replication":"https://pith.science/pith/UOBOWZJYL5AP6DBOR6BCL77IOP/action/replication_record"}},"created_at":"2026-07-05T07:37:57.529939+00:00","updated_at":"2026-07-05T07:37:57.529939+00:00"}