{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UOBOWZJYL5AP6DBOR6BCL77IOP","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"241d053112ac58a11df97ba1dc3ac1a13d503ef0b84fcb732724a889fa721a13","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T19:57:00Z","title_canon_sha256":"7f04d7170b4657d4bddf164ded75aa8c02be0d6289e1776dbe1f0eb5392a3efe"},"schema_version":"1.0","source":{"id":"2401.14490","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.14490","created_at":"2026-07-05T07:37:57Z"},{"alias_kind":"arxiv_version","alias_value":"2401.14490v1","created_at":"2026-07-05T07:37:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14490","created_at":"2026-07-05T07:37:57Z"},{"alias_kind":"pith_short_12","alias_value":"UOBOWZJYL5AP","created_at":"2026-07-05T07:37:57Z"},{"alias_kind":"pith_short_16","alias_value":"UOBOWZJYL5AP6DBO","created_at":"2026-07-05T07:37:57Z"},{"alias_kind":"pith_short_8","alias_value":"UOBOWZJY","created_at":"2026-07-05T07:37:57Z"}],"graph_snapshots":[{"event_id":"sha256:1f1d560040592ce6f98f866d0cd0fcf6bc0d9bd798e0b64dbd9298fe6b142f3d","target":"graph","created_at":"2026-07-05T07:37:57Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2401.14490/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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 ","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","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T19:57:00Z","title":"LongHealth: A Question Answering Benchmark with Long Clinical Documents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14490","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:98d4b1a0e1c2cada4c46959e809cc31e94d4ee2dfd3a6cb830933d557fdb4a28","target":"record","created_at":"2026-07-05T07:37:57Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"241d053112ac58a11df97ba1dc3ac1a13d503ef0b84fcb732724a889fa721a13","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T19:57:00Z","title_canon_sha256":"7f04d7170b4657d4bddf164ded75aa8c02be0d6289e1776dbe1f0eb5392a3efe"},"schema_version":"1.0","source":{"id":"2401.14490","kind":"arxiv","version":1}},"canonical_sha256":"a382eb65385f40ff0c2e8f8225ffe873d02f4be52bb55a225bf565b60aba39fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a382eb65385f40ff0c2e8f8225ffe873d02f4be52bb55a225bf565b60aba39fd","first_computed_at":"2026-07-05T07:37:57.529886Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:37:57.529886Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/kUzVGmiCmiMINDv5IG8ckmEcl03JTsHFnJWEwiOiEhaVwh+WmmeSJoiwEcpNS7q2N5zq+CtlfPppctPw44eCg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:37:57.530367Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.14490","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:98d4b1a0e1c2cada4c46959e809cc31e94d4ee2dfd3a6cb830933d557fdb4a28","sha256:1f1d560040592ce6f98f866d0cd0fcf6bc0d9bd798e0b64dbd9298fe6b142f3d"],"state_sha256":"bb8883401f3b41ba5c8afe1d56fd1e02d2d8963080b4dfee44f350ac5cf6943c"}