{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LCQVOZGEGIL3OLL77GC55J54EK","short_pith_number":"pith:LCQVOZGE","schema_version":"1.0","canonical_sha256":"58a15764c43217b72d7ff985dea7bc2299890c61032258cdef577eb0b362cada","source":{"kind":"arxiv","id":"2605.21496","version":1},"attestation_state":"computed","paper":{"title":"HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Brandon Dent","submitted_at":"2026-04-18T03:40:24Z","abstract_excerpt":"Frontier language models are being deployed into clinical workflows faster than the infrastructure to evaluate them safely. Static medical-QA benchmarks miss the failure modes that matter in emergency medicine: trajectory-level safety collapse, tool misuse, and capitulation under sustained clinical pressure. We present HealthCraft, the first public reinforcement-learning environment that rewards trajectory-level safety under realistic emergency-medicine conditions, adapted from Corecraft. It is built on a FHIR R4 world state with 14 entity types and 3,987 seed entities, exposes 24 MCP tools, a"},"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":"2605.21496","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-18T03:40:24Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"6b4b50de1bf2e4b45824a67d1c235e8a7bf63ef3aa21133752e9740cac97e31b","abstract_canon_sha256":"92d79f1cff7b3951fc0720debbcd1c09e3d19732528b04ca20a1cec57148ac64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T00:02:26.948556Z","signature_b64":"0It5iwLwJKnMCJ7DddqSMA5Hj1bgC1xkN9nb1ZK2kLOL4Y1tCFZ1s+ytvur2PrXe4j8OXvAnHu1rFJDMD5kPAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58a15764c43217b72d7ff985dea7bc2299890c61032258cdef577eb0b362cada","last_reissued_at":"2026-05-22T00:02:26.947987Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T00:02:26.947987Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Brandon Dent","submitted_at":"2026-04-18T03:40:24Z","abstract_excerpt":"Frontier language models are being deployed into clinical workflows faster than the infrastructure to evaluate them safely. Static medical-QA benchmarks miss the failure modes that matter in emergency medicine: trajectory-level safety collapse, tool misuse, and capitulation under sustained clinical pressure. We present HealthCraft, the first public reinforcement-learning environment that rewards trajectory-level safety under realistic emergency-medicine conditions, adapted from Corecraft. It is built on a FHIR R4 world state with 14 entity types and 3,987 seed entities, exposes 24 MCP tools, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21496","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/2605.21496/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":"2605.21496","created_at":"2026-05-22T00:02:26.948050+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21496v1","created_at":"2026-05-22T00:02:26.948050+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21496","created_at":"2026-05-22T00:02:26.948050+00:00"},{"alias_kind":"pith_short_12","alias_value":"LCQVOZGEGIL3","created_at":"2026-05-22T00:02:26.948050+00:00"},{"alias_kind":"pith_short_16","alias_value":"LCQVOZGEGIL3OLL7","created_at":"2026-05-22T00:02:26.948050+00:00"},{"alias_kind":"pith_short_8","alias_value":"LCQVOZGE","created_at":"2026-05-22T00:02:26.948050+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK","json":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK.json","graph_json":"https://pith.science/api/pith-number/LCQVOZGEGIL3OLL77GC55J54EK/graph.json","events_json":"https://pith.science/api/pith-number/LCQVOZGEGIL3OLL77GC55J54EK/events.json","paper":"https://pith.science/paper/LCQVOZGE"},"agent_actions":{"view_html":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK","download_json":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK.json","view_paper":"https://pith.science/paper/LCQVOZGE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21496&json=true","fetch_graph":"https://pith.science/api/pith-number/LCQVOZGEGIL3OLL77GC55J54EK/graph.json","fetch_events":"https://pith.science/api/pith-number/LCQVOZGEGIL3OLL77GC55J54EK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK/action/storage_attestation","attest_author":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK/action/author_attestation","sign_citation":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK/action/citation_signature","submit_replication":"https://pith.science/pith/LCQVOZGEGIL3OLL77GC55J54EK/action/replication_record"}},"created_at":"2026-05-22T00:02:26.948050+00:00","updated_at":"2026-05-22T00:02:26.948050+00:00"}