{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BVQYWRB7TQFAWBFVOMH55OBM3K","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":"86881e6813b830be968f4e74eef32b0b27bd6a8bcc40c424ec7b9bcf43b1153d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T23:04:48Z","title_canon_sha256":"da551d47f09305c7c09151af5e1c76c4527c9d6bdecfe4c6a0b8e6a64a8b053a"},"schema_version":"1.0","source":{"id":"2605.21807","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21807","created_at":"2026-05-22T01:03:33Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21807v1","created_at":"2026-05-22T01:03:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21807","created_at":"2026-05-22T01:03:33Z"},{"alias_kind":"pith_short_12","alias_value":"BVQYWRB7TQFA","created_at":"2026-05-22T01:03:33Z"},{"alias_kind":"pith_short_16","alias_value":"BVQYWRB7TQFAWBFV","created_at":"2026-05-22T01:03:33Z"},{"alias_kind":"pith_short_8","alias_value":"BVQYWRB7","created_at":"2026-05-22T01:03:33Z"}],"graph_snapshots":[{"event_id":"sha256:20ddd02261dca2ccedbf66678bfce8e549a2ab9af3895016334e9472389cdc82","target":"graph","created_at":"2026-05-22T01:03:33Z","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/2605.21807/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Across medical specialties, clinical practice is anchored in evidence-based guidelines that codify best studied diagnostic and treatment pathways. These pathways routinely fall short for the long tail of real-world care not covered by guidelines. Most medical large language models (LLMs), however, are trained to encode common, guideline-focused medical knowledge in their parameters. Current evaluations test models primarily on recalling and reasoning with this memorized content, often in multiple-choice settings. Given the fundamental importance of evidence-based reasoning in medicine, it is n","authors_text":"Andrew Srisuwananukorn, Ashish Manne, Brady Buchanan, Doeun Lee, Frank Wen, James Lim, Kathryn Tobin, Lynda Villagomez, Muge Zhang, Oluwatoba Moninuola, Ping Zhang, Sachin Kumar, Stephen Koesters, Yi Yu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T23:04:48Z","title":"When Cases Get Rare: A Retrieval Benchmark for Off-Guideline Clinical Question Answering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21807","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":"sha256:91069057b698c2dbfdf38e2125ca3e223e8418932c1d8fb3d91410b10bb34ee2","target":"record","created_at":"2026-05-22T01:03:33Z","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":"86881e6813b830be968f4e74eef32b0b27bd6a8bcc40c424ec7b9bcf43b1153d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T23:04:48Z","title_canon_sha256":"da551d47f09305c7c09151af5e1c76c4527c9d6bdecfe4c6a0b8e6a64a8b053a"},"schema_version":"1.0","source":{"id":"2605.21807","kind":"arxiv","version":1}},"canonical_sha256":"0d618b443f9c0a0b04b5730fdeb82cda80502917b5a3ab4361b04545839585be","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0d618b443f9c0a0b04b5730fdeb82cda80502917b5a3ab4361b04545839585be","first_computed_at":"2026-05-22T01:03:33.322526Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:03:33.322526Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HabZeIgi+XB4TUo3S5GQO1Kd/8q/QPyapQKQwXRNFcawisEfzeAdHLD+n3MS4Nh/a5CSDpfkCAPa6nnKa89RBA==","signature_status":"signed_v1","signed_at":"2026-05-22T01:03:33.323032Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.21807","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:91069057b698c2dbfdf38e2125ca3e223e8418932c1d8fb3d91410b10bb34ee2","sha256:20ddd02261dca2ccedbf66678bfce8e549a2ab9af3895016334e9472389cdc82"],"state_sha256":"0654f57cd2fc619595db56bf1162c3c9d316f8d17cb7e2a3bc09715c13b52c07"}