{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PMTOWHSIY7M5TB3GXCMHIA7Z4M","short_pith_number":"pith:PMTOWHSI","schema_version":"1.0","canonical_sha256":"7b26eb1e48c7d9d98766b8987403f9e330759de9947f1fc83cd9d1e7b2b938e9","source":{"kind":"arxiv","id":"1902.07208","version":3},"attestation_state":"computed","paper":{"title":"Transfusion: Understanding Transfer Learning for Medical Imaging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CV","authors_text":"Chiyuan Zhang, Jon Kleinberg, Maithra Raghu, Samy Bengio","submitted_at":"2019-02-14T01:54:53Z","abstract_excerpt":"Transfer learning from natural image datasets, particularly ImageNet, using standard large models and corresponding pretrained weights has become a de-facto method for deep learning applications to medical imaging. However, there are fundamental differences in data sizes, features and task specifications between natural image classification and the target medical tasks, and there is little understanding of the effects of transfer. In this paper, we explore properties of transfer learning for medical imaging. A performance evaluation on two large scale medical imaging tasks shows that surprisin"},"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":"1902.07208","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-02-14T01:54:53Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"87f7e8da2d1908017338d433fb48374838fc9bd7fb4067cabc3b965fac11a508","abstract_canon_sha256":"b1256bbce9013df869c36fd3310838662219bde5e909a3e325a33923b6af6942"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:15:47.027398Z","signature_b64":"RnjmSFynoktSSeXgn8ITitOlFdYhkloXwLiygs27Pmi0IVWC3rBqnV2xQzPYEBsClAer8q55+cK+aHpviwzVDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b26eb1e48c7d9d98766b8987403f9e330759de9947f1fc83cd9d1e7b2b938e9","last_reissued_at":"2026-07-05T00:15:47.026865Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:15:47.026865Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transfusion: Understanding Transfer Learning for Medical Imaging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CV","authors_text":"Chiyuan Zhang, Jon Kleinberg, Maithra Raghu, Samy Bengio","submitted_at":"2019-02-14T01:54:53Z","abstract_excerpt":"Transfer learning from natural image datasets, particularly ImageNet, using standard large models and corresponding pretrained weights has become a de-facto method for deep learning applications to medical imaging. However, there are fundamental differences in data sizes, features and task specifications between natural image classification and the target medical tasks, and there is little understanding of the effects of transfer. In this paper, we explore properties of transfer learning for medical imaging. A performance evaluation on two large scale medical imaging tasks shows that surprisin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.07208","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/1902.07208/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":"1902.07208","created_at":"2026-07-05T00:15:47.026936+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.07208v3","created_at":"2026-07-05T00:15:47.026936+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.07208","created_at":"2026-07-05T00:15:47.026936+00:00"},{"alias_kind":"pith_short_12","alias_value":"PMTOWHSIY7M5","created_at":"2026-07-05T00:15:47.026936+00:00"},{"alias_kind":"pith_short_16","alias_value":"PMTOWHSIY7M5TB3G","created_at":"2026-07-05T00:15:47.026936+00:00"},{"alias_kind":"pith_short_8","alias_value":"PMTOWHSI","created_at":"2026-07-05T00:15:47.026936+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18707","citing_title":"PEFT-MedSAM: Efficient Fine-Tuning of Medical Foundation Models for Explainable Skin Lesion Segmentation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2506.13015","citing_title":"Geometric Embedding Alignment via Curvature Matching in Transfer Learning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"1910.04867","citing_title":"A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16588","citing_title":"Vision Transformers Need Registers","ref_index":238,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21927","citing_title":"Fine-Tuning Regimes Define Distinct Continual Learning Problems","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M","json":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M.json","graph_json":"https://pith.science/api/pith-number/PMTOWHSIY7M5TB3GXCMHIA7Z4M/graph.json","events_json":"https://pith.science/api/pith-number/PMTOWHSIY7M5TB3GXCMHIA7Z4M/events.json","paper":"https://pith.science/paper/PMTOWHSI"},"agent_actions":{"view_html":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M","download_json":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M.json","view_paper":"https://pith.science/paper/PMTOWHSI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.07208&json=true","fetch_graph":"https://pith.science/api/pith-number/PMTOWHSIY7M5TB3GXCMHIA7Z4M/graph.json","fetch_events":"https://pith.science/api/pith-number/PMTOWHSIY7M5TB3GXCMHIA7Z4M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M/action/storage_attestation","attest_author":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M/action/author_attestation","sign_citation":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M/action/citation_signature","submit_replication":"https://pith.science/pith/PMTOWHSIY7M5TB3GXCMHIA7Z4M/action/replication_record"}},"created_at":"2026-07-05T00:15:47.026936+00:00","updated_at":"2026-07-05T00:15:47.026936+00:00"}