{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KLWD2QD35Y5JNTDTZUU4PTZU2Y","short_pith_number":"pith:KLWD2QD3","schema_version":"1.0","canonical_sha256":"52ec3d407bee3a96cc73cd29c7cf34d6148de3ef068e7614dfe409afc5f14190","source":{"kind":"arxiv","id":"2312.12865","version":3},"attestation_state":"computed","paper":{"title":"RadEdit: stress-testing biomedical vision models via diffusion image editing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Aditya Nori, Boris van Breugel, Daniel C. Castro, Fernando P\\'erez-Garc\\'ia, Hannah Richardson, Harshita Sharma, Javier Alvarez-Valle, Maria T. A. Wetscherek, Matthew P. Lungren, Maximilian Ilse, Ozan Oktay, Pedro P. Sanchez, Sam Bond-Taylor, Valentina Salvatelli","submitted_at":"2023-12-20T09:27:41Z","abstract_excerpt":"Biomedical imaging datasets are often small and biased, meaning that real-world performance of predictive models can be substantially lower than expected from internal testing. This work proposes using generative image editing to simulate dataset shifts and diagnose failure modes of biomedical vision models; this can be used in advance of deployment to assess readiness, potentially reducing cost and patient harm. Existing editing methods can produce undesirable changes, with spurious correlations learned due to the co-occurrence of disease and treatment interventions, limiting practical applic"},"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":"2312.12865","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-12-20T09:27:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f1912141dd1ff6f8ffc3516ac14a4d08601ee99a95f85cd6d5cfce2439938860","abstract_canon_sha256":"9599976dca7b0eae4aca371e9d32fa82ab98c1b60199513ed3c5d1f18de864f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:40.788332Z","signature_b64":"AEW4lGEQqhz2bZIlzuC1NTFHUPRBBJeTgqyyhid0EXkjbs1IGJ1S41tgPiwEkN2q3+Sqb2WZg7mMAhXi/EM8CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52ec3d407bee3a96cc73cd29c7cf34d6148de3ef068e7614dfe409afc5f14190","last_reissued_at":"2026-07-05T09:49:40.787873Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:40.787873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RadEdit: stress-testing biomedical vision models via diffusion image editing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Aditya Nori, Boris van Breugel, Daniel C. Castro, Fernando P\\'erez-Garc\\'ia, Hannah Richardson, Harshita Sharma, Javier Alvarez-Valle, Maria T. A. Wetscherek, Matthew P. Lungren, Maximilian Ilse, Ozan Oktay, Pedro P. Sanchez, Sam Bond-Taylor, Valentina Salvatelli","submitted_at":"2023-12-20T09:27:41Z","abstract_excerpt":"Biomedical imaging datasets are often small and biased, meaning that real-world performance of predictive models can be substantially lower than expected from internal testing. This work proposes using generative image editing to simulate dataset shifts and diagnose failure modes of biomedical vision models; this can be used in advance of deployment to assess readiness, potentially reducing cost and patient harm. Existing editing methods can produce undesirable changes, with spurious correlations learned due to the co-occurrence of disease and treatment interventions, limiting practical applic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.12865","kind":"arxiv","version":3},"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/2312.12865/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":"2312.12865","created_at":"2026-07-05T09:49:40.787922+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.12865v3","created_at":"2026-07-05T09:49:40.787922+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.12865","created_at":"2026-07-05T09:49:40.787922+00:00"},{"alias_kind":"pith_short_12","alias_value":"KLWD2QD35Y5J","created_at":"2026-07-05T09:49:40.787922+00:00"},{"alias_kind":"pith_short_16","alias_value":"KLWD2QD35Y5JNTDT","created_at":"2026-07-05T09:49:40.787922+00:00"},{"alias_kind":"pith_short_8","alias_value":"KLWD2QD3","created_at":"2026-07-05T09:49:40.787922+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.07883","citing_title":"Diffusion Counterfactual Generation with Semantic Abduction","ref_index":81,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y","json":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y.json","graph_json":"https://pith.science/api/pith-number/KLWD2QD35Y5JNTDTZUU4PTZU2Y/graph.json","events_json":"https://pith.science/api/pith-number/KLWD2QD35Y5JNTDTZUU4PTZU2Y/events.json","paper":"https://pith.science/paper/KLWD2QD3"},"agent_actions":{"view_html":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y","download_json":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y.json","view_paper":"https://pith.science/paper/KLWD2QD3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.12865&json=true","fetch_graph":"https://pith.science/api/pith-number/KLWD2QD35Y5JNTDTZUU4PTZU2Y/graph.json","fetch_events":"https://pith.science/api/pith-number/KLWD2QD35Y5JNTDTZUU4PTZU2Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y/action/storage_attestation","attest_author":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y/action/author_attestation","sign_citation":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y/action/citation_signature","submit_replication":"https://pith.science/pith/KLWD2QD35Y5JNTDTZUU4PTZU2Y/action/replication_record"}},"created_at":"2026-07-05T09:49:40.787922+00:00","updated_at":"2026-07-05T09:49:40.787922+00:00"}