{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:SI5TMZ4TNSGBW6HJA63N7U3YVQ","short_pith_number":"pith:SI5TMZ4T","canonical_record":{"source":{"id":"2607.18828","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-21T08:05:35Z","cross_cats_sorted":[],"title_canon_sha256":"a47613bff8b5478abd117e44b0b022531a15fc118b531ab4affe1045f079529b","abstract_canon_sha256":"6edad88a3d4ba1c072d50aeeb404efe73e9901d14d7134618ae34f2ab0842a24"},"schema_version":"1.0"},"canonical_sha256":"923b3667936c8c1b78e907b6dfd378ac1db740c5fbdbadbb6fd91244197466d0","source":{"kind":"arxiv","id":"2607.18828","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.18828","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"arxiv_version","alias_value":"2607.18828v1","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.18828","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"pith_short_12","alias_value":"SI5TMZ4TNSGB","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"pith_short_16","alias_value":"SI5TMZ4TNSGBW6HJ","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"pith_short_8","alias_value":"SI5TMZ4T","created_at":"2026-07-22T01:23:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:SI5TMZ4TNSGBW6HJA63N7U3YVQ","target":"record","payload":{"canonical_record":{"source":{"id":"2607.18828","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-21T08:05:35Z","cross_cats_sorted":[],"title_canon_sha256":"a47613bff8b5478abd117e44b0b022531a15fc118b531ab4affe1045f079529b","abstract_canon_sha256":"6edad88a3d4ba1c072d50aeeb404efe73e9901d14d7134618ae34f2ab0842a24"},"schema_version":"1.0"},"canonical_sha256":"923b3667936c8c1b78e907b6dfd378ac1db740c5fbdbadbb6fd91244197466d0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-22T01:23:14.741760Z","signature_b64":"mMaiWAcG+Z8Czgd2pcdQApw5u1GKmtDwL49AvcpzAqeTIzTABB8HlzVs16NZ3arpf0LQKkisCsOdE9aV3LwgCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"923b3667936c8c1b78e907b6dfd378ac1db740c5fbdbadbb6fd91244197466d0","last_reissued_at":"2026-07-22T01:23:14.740847Z","signature_status":"signed_v1","first_computed_at":"2026-07-22T01:23:14.740847Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.18828","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-22T01:23:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p8lhGWl9zU3dxuICVY9Qr9C8/uEOPDd5yPozJCpXfsmihwSJmUsCeECAfnFCGEdYcTkQKRtfNYn+O8h2cQTXAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T04:36:12.698821Z"},"content_sha256":"4e31dff8919d209a1dd5910789a891e4df2492fcf69376760eb26e0332fc2ce5","schema_version":"1.0","event_id":"sha256:4e31dff8919d209a1dd5910789a891e4df2492fcf69376760eb26e0332fc2ce5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:SI5TMZ4TNSGBW6HJA63N7U3YVQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating medical AI under missing information: same-provider judges and human raters change apparent safety","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Koyar Afrasyab","submitted_at":"2026-07-21T08:05:35Z","abstract_excerpt":"Readiness stress-testing of medical AI has focused on closed-ended and multimodal benchmarks. We extend it to open-ended clinical conversation under missing information, where safe behavior means recognizing absent information and qualifying, clarifying, or not over-committing - and where the evaluator becomes part of the measurement. We stress-test four models - three flagships (Claude Opus 4.8, GPT-5.5, Grok 4.3) and one mid-tier model (Gemini 3.5 Flash) - by deleting the latter half of the final user turn in HealthBench conversations, grading responses with a four-provider LLM-judge panel a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.18828","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/2607.18828/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-22T01:23:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"04y7qCrNG0812YxLksC8wrPLkVelUTfejX8keLTK2CRZTW8vDFFdUnMBcsVG1iuWqM4KRh5hGuyDkxLbL3U5AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T04:36:12.699320Z"},"content_sha256":"2f81bb966719c09badf76ed5de94cd01538419a6a058608c663a3d81088c6c46","schema_version":"1.0","event_id":"sha256:2f81bb966719c09badf76ed5de94cd01538419a6a058608c663a3d81088c6c46"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ/bundle.json","state_url":"https://pith.science/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-20T04:36:12Z","links":{"resolver":"https://pith.science/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ","bundle":"https://pith.science/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ/bundle.json","state":"https://pith.science/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SI5TMZ4TNSGBW6HJA63N7U3YVQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SI5TMZ4TNSGBW6HJA63N7U3YVQ","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":"6edad88a3d4ba1c072d50aeeb404efe73e9901d14d7134618ae34f2ab0842a24","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-21T08:05:35Z","title_canon_sha256":"a47613bff8b5478abd117e44b0b022531a15fc118b531ab4affe1045f079529b"},"schema_version":"1.0","source":{"id":"2607.18828","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.18828","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"arxiv_version","alias_value":"2607.18828v1","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.18828","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"pith_short_12","alias_value":"SI5TMZ4TNSGB","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"pith_short_16","alias_value":"SI5TMZ4TNSGBW6HJ","created_at":"2026-07-22T01:23:14Z"},{"alias_kind":"pith_short_8","alias_value":"SI5TMZ4T","created_at":"2026-07-22T01:23:14Z"}],"graph_snapshots":[{"event_id":"sha256:2f81bb966719c09badf76ed5de94cd01538419a6a058608c663a3d81088c6c46","target":"graph","created_at":"2026-07-22T01:23:14Z","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/2607.18828/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Readiness stress-testing of medical AI has focused on closed-ended and multimodal benchmarks. We extend it to open-ended clinical conversation under missing information, where safe behavior means recognizing absent information and qualifying, clarifying, or not over-committing - and where the evaluator becomes part of the measurement. We stress-test four models - three flagships (Claude Opus 4.8, GPT-5.5, Grok 4.3) and one mid-tier model (Gemini 3.5 Flash) - by deleting the latter half of the final user turn in HealthBench conversations, grading responses with a four-provider LLM-judge panel a","authors_text":"Koyar Afrasyab","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-21T08:05:35Z","title":"Evaluating medical AI under missing information: same-provider judges and human raters change apparent safety"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.18828","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":"sha256:4e31dff8919d209a1dd5910789a891e4df2492fcf69376760eb26e0332fc2ce5","target":"record","created_at":"2026-07-22T01:23:14Z","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":"6edad88a3d4ba1c072d50aeeb404efe73e9901d14d7134618ae34f2ab0842a24","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-21T08:05:35Z","title_canon_sha256":"a47613bff8b5478abd117e44b0b022531a15fc118b531ab4affe1045f079529b"},"schema_version":"1.0","source":{"id":"2607.18828","kind":"arxiv","version":1}},"canonical_sha256":"923b3667936c8c1b78e907b6dfd378ac1db740c5fbdbadbb6fd91244197466d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"923b3667936c8c1b78e907b6dfd378ac1db740c5fbdbadbb6fd91244197466d0","first_computed_at":"2026-07-22T01:23:14.740847Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-22T01:23:14.740847Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mMaiWAcG+Z8Czgd2pcdQApw5u1GKmtDwL49AvcpzAqeTIzTABB8HlzVs16NZ3arpf0LQKkisCsOdE9aV3LwgCA==","signature_status":"signed_v1","signed_at":"2026-07-22T01:23:14.741760Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.18828","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4e31dff8919d209a1dd5910789a891e4df2492fcf69376760eb26e0332fc2ce5","sha256:2f81bb966719c09badf76ed5de94cd01538419a6a058608c663a3d81088c6c46"],"state_sha256":"89bf34e88de1daa2741c22f512e7a44e70f407fa1414a1fb9066cf31f94aae29"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S9qp7RwLpbvqxFmx8ydeEerjnBtGYLrWTSp4i05d7xBsvuZ67vgZYkTvGIbHZivGglcamxvqGqgYi5+2NojiCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T04:36:12.707086Z","bundle_sha256":"f4fd8f41f5e5703c8559463a1f797a9cd4e295e4633f4424e1776f179b49b104"}}