{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RY53GOTEC6L24RBXCVN6D2QU5H","short_pith_number":"pith:RY53GOTE","schema_version":"1.0","canonical_sha256":"8e3bb33a641797ae4437155be1ea14e9ef68c1fb264b3715823e52b814087921","source":{"kind":"arxiv","id":"2505.14963","version":1},"attestation_state":"computed","paper":{"title":"MedBrowseComp: Benchmarking Medical Deep Research and Computer Use","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danielle S. Bitterman, Hugo Aerts, Jack Gallifant, Jeremy Warner, Pedro Moreira, Sam Schmidgall, Shan Chen, Thomas Hartvigsen, Yuxin Xiao","submitted_at":"2025-05-20T22:42:33Z","abstract_excerpt":"Large language models (LLMs) are increasingly envisioned as decision-support tools in clinical practice, yet safe clinical reasoning demands integrating heterogeneous knowledge bases -- trials, primary studies, regulatory documents, and cost data -- under strict accuracy constraints. Existing evaluations often rely on synthetic prompts, reduce the task to single-hop factoid queries, or conflate reasoning with open-ended generation, leaving their real-world utility unclear. To close this gap, we present MedBrowseComp, the first benchmark that systematically tests an agent's ability to reliably "},"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":"2505.14963","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T22:42:33Z","cross_cats_sorted":[],"title_canon_sha256":"9f704fde3ee58d346fc0604b2d0e293aafd786bb80eb6759e710818ee07b958b","abstract_canon_sha256":"7a393126dc4b254f60c0b5ad98a11939cc6a7205617b1e9fb39d40903244a2e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:38.412402Z","signature_b64":"aZSry6zr93gXd+9hDIkEzgrALeTv9izSn19/Fg8mSYBWz7stL6ss9m+B4gaPpUbG3pmEu/QnpHFPVDwO1ouECQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e3bb33a641797ae4437155be1ea14e9ef68c1fb264b3715823e52b814087921","last_reissued_at":"2026-07-05T11:06:38.411908Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:38.411908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MedBrowseComp: Benchmarking Medical Deep Research and Computer Use","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danielle S. Bitterman, Hugo Aerts, Jack Gallifant, Jeremy Warner, Pedro Moreira, Sam Schmidgall, Shan Chen, Thomas Hartvigsen, Yuxin Xiao","submitted_at":"2025-05-20T22:42:33Z","abstract_excerpt":"Large language models (LLMs) are increasingly envisioned as decision-support tools in clinical practice, yet safe clinical reasoning demands integrating heterogeneous knowledge bases -- trials, primary studies, regulatory documents, and cost data -- under strict accuracy constraints. Existing evaluations often rely on synthetic prompts, reduce the task to single-hop factoid queries, or conflate reasoning with open-ended generation, leaving their real-world utility unclear. To close this gap, we present MedBrowseComp, the first benchmark that systematically tests an agent's ability to reliably "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14963","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/2505.14963/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":"2505.14963","created_at":"2026-07-05T11:06:38.411967+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14963v1","created_at":"2026-07-05T11:06:38.411967+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14963","created_at":"2026-07-05T11:06:38.411967+00:00"},{"alias_kind":"pith_short_12","alias_value":"RY53GOTEC6L2","created_at":"2026-07-05T11:06:38.411967+00:00"},{"alias_kind":"pith_short_16","alias_value":"RY53GOTEC6L24RBX","created_at":"2026-07-05T11:06:38.411967+00:00"},{"alias_kind":"pith_short_8","alias_value":"RY53GOTE","created_at":"2026-07-05T11:06:38.411967+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06374","citing_title":"VaseMuseum: Digital Intelligent Museum for Ancient Greek Pottery","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20683","citing_title":"From Question Answering to Task Completion: A Survey on Agent System and Harness Design","ref_index":206,"is_internal_anchor":false},{"citing_arxiv_id":"2507.12261","citing_title":"Infherno: End-to-end Agent-based FHIR Resource Synthesis from Free-form Clinical Notes","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2602.12705","citing_title":"MedXIAOHE: A Comprehensive Recipe for Building Medical MLLMs","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H","json":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H.json","graph_json":"https://pith.science/api/pith-number/RY53GOTEC6L24RBXCVN6D2QU5H/graph.json","events_json":"https://pith.science/api/pith-number/RY53GOTEC6L24RBXCVN6D2QU5H/events.json","paper":"https://pith.science/paper/RY53GOTE"},"agent_actions":{"view_html":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H","download_json":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H.json","view_paper":"https://pith.science/paper/RY53GOTE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14963&json=true","fetch_graph":"https://pith.science/api/pith-number/RY53GOTEC6L24RBXCVN6D2QU5H/graph.json","fetch_events":"https://pith.science/api/pith-number/RY53GOTEC6L24RBXCVN6D2QU5H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H/action/storage_attestation","attest_author":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H/action/author_attestation","sign_citation":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H/action/citation_signature","submit_replication":"https://pith.science/pith/RY53GOTEC6L24RBXCVN6D2QU5H/action/replication_record"}},"created_at":"2026-07-05T11:06:38.411967+00:00","updated_at":"2026-07-05T11:06:38.411967+00:00"}