{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TAXHFACXOXZHYT3TLPELXIVANA","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":"01b4c2d32c3c818b9760876afea3d082116af6b94b7b44402b487b689324fa6b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-10T15:38:44Z","title_canon_sha256":"65078ee46de330ed2af33f2f61d41efc29fe3aa4720b77d17b503f6a09e09200"},"schema_version":"1.0","source":{"id":"2503.07459","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.07459","created_at":"2026-06-19T16:10:29Z"},{"alias_kind":"arxiv_version","alias_value":"2503.07459v3","created_at":"2026-06-19T16:10:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.07459","created_at":"2026-06-19T16:10:29Z"},{"alias_kind":"pith_short_12","alias_value":"TAXHFACXOXZH","created_at":"2026-06-19T16:10:29Z"},{"alias_kind":"pith_short_16","alias_value":"TAXHFACXOXZHYT3T","created_at":"2026-06-19T16:10:29Z"},{"alias_kind":"pith_short_8","alias_value":"TAXHFACX","created_at":"2026-06-19T16:10:29Z"}],"graph_snapshots":[{"event_id":"sha256:cee803663a222428278fb5d5447c2a685be6189371d2a85a0e351950ced3e11c","target":"graph","created_at":"2026-06-19T16:10:29Z","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/2503.07459/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Complex medical reasoning requires integrating heterogeneous clinical evidence across multiple inference steps. Large language models (LLMs) now approach this through two routes: internalized reasoning and externalized agent scaffolding (frameworks that decompose problems collaboratively amongst multiple LLMs). To determine whether these routes are exclusive or complementary, we introduce MedicalAgentsBench, a filtered benchmark of 862 complex clinical questions drawn from the union of eight medical datasets via difficulty-aware curation and contamination screening. Evaluating three internaliz","authors_text":"Arman Cohan, Chenglin Wu, Fang Wu, Jiapeng Chen, Jiayi Zhang, Jinyu Xiang, Jiwoong Sohn, Mark Gerstein, Wenqi Shi, Xiangru Tang, Yanjun Shao, Yilun Zhao, Yuxuan Liao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-10T15:38:44Z","title":"MedicalAgentsBench for Complex Medical Reasoning: Comparing Internalized Reasoning Models versus Externalized Agent-based Frameworks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.07459","kind":"arxiv","version":3},"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":"sha256:465074bcbd624872bbee27a76e8104a6947a43f8b537244b5f641379d26dc97c","target":"record","created_at":"2026-06-19T16:10:29Z","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":"01b4c2d32c3c818b9760876afea3d082116af6b94b7b44402b487b689324fa6b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-10T15:38:44Z","title_canon_sha256":"65078ee46de330ed2af33f2f61d41efc29fe3aa4720b77d17b503f6a09e09200"},"schema_version":"1.0","source":{"id":"2503.07459","kind":"arxiv","version":3}},"canonical_sha256":"982e72805775f27c4f735bc8bba2a0680b5fbe5c1b82f2225a3f49d7c3008865","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"982e72805775f27c4f735bc8bba2a0680b5fbe5c1b82f2225a3f49d7c3008865","first_computed_at":"2026-06-19T16:10:29.419374Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:10:29.419374Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rAb8yceQXEuD0nfIoJ6cvTSIcL9Zktb1Koak0b5tvN989E0VNusWhdOAYl8xIlcWU1mXNI08DGnPnoT8acWNAw==","signature_status":"signed_v1","signed_at":"2026-06-19T16:10:29.419800Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.07459","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:465074bcbd624872bbee27a76e8104a6947a43f8b537244b5f641379d26dc97c","sha256:cee803663a222428278fb5d5447c2a685be6189371d2a85a0e351950ced3e11c"],"state_sha256":"05539c77c0bf9704816653e6d1d61f622a314a1bd8626f22385e3e154d481165"}