{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4NAAQKLOHCUSOQJ3NMLSLHS6TN","short_pith_number":"pith:4NAAQKLO","canonical_record":{"source":{"id":"2507.05755","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-08T07:58:52Z","cross_cats_sorted":[],"title_canon_sha256":"0f9ce79aa446d9ac9b1425f114ffe122d1e0b0f63d588075556ae4a0dd18e655","abstract_canon_sha256":"425561475a3c4ee46d5180258528914f8551e01733b89daa63e72b821b486db5"},"schema_version":"1.0"},"canonical_sha256":"e34008296e38a927413b6b17259e5e9b54b5d450c689d60cea32777175f6d4fe","source":{"kind":"arxiv","id":"2507.05755","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.05755","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"arxiv_version","alias_value":"2507.05755v1","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.05755","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"pith_short_12","alias_value":"4NAAQKLOHCUS","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"pith_short_16","alias_value":"4NAAQKLOHCUSOQJ3","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"pith_short_8","alias_value":"4NAAQKLO","created_at":"2026-07-05T11:33:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4NAAQKLOHCUSOQJ3NMLSLHS6TN","target":"record","payload":{"canonical_record":{"source":{"id":"2507.05755","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-08T07:58:52Z","cross_cats_sorted":[],"title_canon_sha256":"0f9ce79aa446d9ac9b1425f114ffe122d1e0b0f63d588075556ae4a0dd18e655","abstract_canon_sha256":"425561475a3c4ee46d5180258528914f8551e01733b89daa63e72b821b486db5"},"schema_version":"1.0"},"canonical_sha256":"e34008296e38a927413b6b17259e5e9b54b5d450c689d60cea32777175f6d4fe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:35.876054Z","signature_b64":"V6c47HblXnX0fR+SQpqV0sE6+gfJUJsEiALWeiBTDhElVGwcSPtUXtgvi3x3UdrlyzbeHp+Llu6iDrKuM5NWCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e34008296e38a927413b6b17259e5e9b54b5d450c689d60cea32777175f6d4fe","last_reissued_at":"2026-07-05T11:33:35.875611Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:35.875611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.05755","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":"2026-07-05T11:33:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9sK4+68J/crN65U1xzL/AdJhboOg7E5nLUn/KqHy5twlrunbvEksHiiMNw2+dvJcx/aQ5uELk4uFOEAAuxMhCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:47:32.796229Z"},"content_sha256":"30c53e3ada6c2c7d1c71a9ff60a2e8f4a019b2e6733aab1eeed44fda933d807a","schema_version":"1.0","event_id":"sha256:30c53e3ada6c2c7d1c71a9ff60a2e8f4a019b2e6733aab1eeed44fda933d807a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4NAAQKLOHCUSOQJ3NMLSLHS6TN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An autonomous agent for auditing and improving the reliability of clinical AI models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Florian Buettner, Lukas Kuhn","submitted_at":"2025-07-08T07:58:52Z","abstract_excerpt":"The deployment of AI models in clinical practice faces a critical challenge: models achieving expert-level performance on benchmarks can fail catastrophically when confronted with real-world variations in medical imaging. Minor shifts in scanner hardware, lighting or demographics can erode accuracy, but currently reliability auditing to identify such catastrophic failure cases before deployment is a bespoke and time-consuming process. Practitioners lack accessible and interpretable tools to expose and repair hidden failure modes. Here we introduce ModelAuditor, a self-reflective agent that con"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.05755","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/2507.05755/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":"2026-07-05T11:33:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"psmy9BHe9TWyDd9PNRwbkRwfQhACqY8nW0puCNdJzWte6XEAcw/ebDxVXzC4yHt8vro46CZ6ky795aVZn2dwCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:47:32.796731Z"},"content_sha256":"1442e3a6a44a4b933d7336fc0016bfe612f21f7c850fd9e569d54e720246ea2a","schema_version":"1.0","event_id":"sha256:1442e3a6a44a4b933d7336fc0016bfe612f21f7c850fd9e569d54e720246ea2a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN/bundle.json","state_url":"https://pith.science/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN/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":"2026-08-08T11:47:32Z","links":{"resolver":"https://pith.science/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN","bundle":"https://pith.science/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN/bundle.json","state":"https://pith.science/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4NAAQKLOHCUSOQJ3NMLSLHS6TN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4NAAQKLOHCUSOQJ3NMLSLHS6TN","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":"425561475a3c4ee46d5180258528914f8551e01733b89daa63e72b821b486db5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-08T07:58:52Z","title_canon_sha256":"0f9ce79aa446d9ac9b1425f114ffe122d1e0b0f63d588075556ae4a0dd18e655"},"schema_version":"1.0","source":{"id":"2507.05755","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.05755","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"arxiv_version","alias_value":"2507.05755v1","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.05755","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"pith_short_12","alias_value":"4NAAQKLOHCUS","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"pith_short_16","alias_value":"4NAAQKLOHCUSOQJ3","created_at":"2026-07-05T11:33:35Z"},{"alias_kind":"pith_short_8","alias_value":"4NAAQKLO","created_at":"2026-07-05T11:33:35Z"}],"graph_snapshots":[{"event_id":"sha256:1442e3a6a44a4b933d7336fc0016bfe612f21f7c850fd9e569d54e720246ea2a","target":"graph","created_at":"2026-07-05T11:33:35Z","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/2507.05755/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The deployment of AI models in clinical practice faces a critical challenge: models achieving expert-level performance on benchmarks can fail catastrophically when confronted with real-world variations in medical imaging. Minor shifts in scanner hardware, lighting or demographics can erode accuracy, but currently reliability auditing to identify such catastrophic failure cases before deployment is a bespoke and time-consuming process. Practitioners lack accessible and interpretable tools to expose and repair hidden failure modes. Here we introduce ModelAuditor, a self-reflective agent that con","authors_text":"Florian Buettner, Lukas Kuhn","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-08T07:58:52Z","title":"An autonomous agent for auditing and improving the reliability of clinical AI models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.05755","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":"sha256:30c53e3ada6c2c7d1c71a9ff60a2e8f4a019b2e6733aab1eeed44fda933d807a","target":"record","created_at":"2026-07-05T11:33:35Z","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":"425561475a3c4ee46d5180258528914f8551e01733b89daa63e72b821b486db5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-08T07:58:52Z","title_canon_sha256":"0f9ce79aa446d9ac9b1425f114ffe122d1e0b0f63d588075556ae4a0dd18e655"},"schema_version":"1.0","source":{"id":"2507.05755","kind":"arxiv","version":1}},"canonical_sha256":"e34008296e38a927413b6b17259e5e9b54b5d450c689d60cea32777175f6d4fe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e34008296e38a927413b6b17259e5e9b54b5d450c689d60cea32777175f6d4fe","first_computed_at":"2026-07-05T11:33:35.875611Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:33:35.875611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"V6c47HblXnX0fR+SQpqV0sE6+gfJUJsEiALWeiBTDhElVGwcSPtUXtgvi3x3UdrlyzbeHp+Llu6iDrKuM5NWCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:33:35.876054Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.05755","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:30c53e3ada6c2c7d1c71a9ff60a2e8f4a019b2e6733aab1eeed44fda933d807a","sha256:1442e3a6a44a4b933d7336fc0016bfe612f21f7c850fd9e569d54e720246ea2a"],"state_sha256":"4bfb11cbd031122d0590691025cb5ca8187417628122be418411cdeed1becb32"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/bOt3k1fMjX84RhBwOBNVsHZ/IpCF36KE+szVvYBeS3aI2j9XTAvlzC3QRupqev639KGFn5L4wllAAEbBx50Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T11:47:32.800465Z","bundle_sha256":"0eb103f3ce30df1e6d78512db5de8e2ee6b906ecbecf4d1bdac2c70184bd98dc"}}