{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:6DUBUA4WDPQEGG47BJ7YOROKVJ","short_pith_number":"pith:6DUBUA4W","schema_version":"1.0","canonical_sha256":"f0e81a03961be0431b9f0a7f8745caaa6ec90ca74a7477824daddf5bbcf46a2e","source":{"kind":"arxiv","id":"2008.06775","version":1},"attestation_state":"computed","paper":{"title":"Model Patching: Closing the Subgroup Performance Gap with Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Albert Gu, Christopher R\\'e, Karan Goel, Yixuan Li","submitted_at":"2020-08-15T20:01:23Z","abstract_excerpt":"Classifiers in machine learning are often brittle when deployed. Particularly concerning are models with inconsistent performance on specific subgroups of a class, e.g., exhibiting disparities in skin cancer classification in the presence or absence of a spurious bandage. To mitigate these performance differences, we introduce model patching, a two-stage framework for improving robustness that encourages the model to be invariant to subgroup differences, and focus on class information shared by subgroups. Model patching first models subgroup features within a class and learns semantic transfor"},"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":"2008.06775","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-08-15T20:01:23Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"3dacb670e0a7159ea26402424049452484262fad7a4d9704eb693b65ab8e0a51","abstract_canon_sha256":"21c8abb83c12ff0cb8fca3a5193d92f63ca8411e047cf136cc0295f4dfedb35f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:27:32.095750Z","signature_b64":"M5/VXqf/mxUCtG13awGI5y1jzbRp4WuFg98XAyV0v/3GdNc/fmW74MrDykuR8NYuhHDkLJ49RqrK5v34gNpUBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f0e81a03961be0431b9f0a7f8745caaa6ec90ca74a7477824daddf5bbcf46a2e","last_reissued_at":"2026-07-05T01:27:32.095274Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:27:32.095274Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model Patching: Closing the Subgroup Performance Gap with Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Albert Gu, Christopher R\\'e, Karan Goel, Yixuan Li","submitted_at":"2020-08-15T20:01:23Z","abstract_excerpt":"Classifiers in machine learning are often brittle when deployed. Particularly concerning are models with inconsistent performance on specific subgroups of a class, e.g., exhibiting disparities in skin cancer classification in the presence or absence of a spurious bandage. To mitigate these performance differences, we introduce model patching, a two-stage framework for improving robustness that encourages the model to be invariant to subgroup differences, and focus on class information shared by subgroups. Model patching first models subgroup features within a class and learns semantic transfor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.06775","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/2008.06775/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":"2008.06775","created_at":"2026-07-05T01:27:32.095335+00:00"},{"alias_kind":"arxiv_version","alias_value":"2008.06775v1","created_at":"2026-07-05T01:27:32.095335+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.06775","created_at":"2026-07-05T01:27:32.095335+00:00"},{"alias_kind":"pith_short_12","alias_value":"6DUBUA4WDPQE","created_at":"2026-07-05T01:27:32.095335+00:00"},{"alias_kind":"pith_short_16","alias_value":"6DUBUA4WDPQEGG47","created_at":"2026-07-05T01:27:32.095335+00:00"},{"alias_kind":"pith_short_8","alias_value":"6DUBUA4W","created_at":"2026-07-05T01:27:32.095335+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.02719","citing_title":"MOMO: Mars Orbital Model Foundation Model for Mars Orbital Applications","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ","json":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ.json","graph_json":"https://pith.science/api/pith-number/6DUBUA4WDPQEGG47BJ7YOROKVJ/graph.json","events_json":"https://pith.science/api/pith-number/6DUBUA4WDPQEGG47BJ7YOROKVJ/events.json","paper":"https://pith.science/paper/6DUBUA4W"},"agent_actions":{"view_html":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ","download_json":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ.json","view_paper":"https://pith.science/paper/6DUBUA4W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2008.06775&json=true","fetch_graph":"https://pith.science/api/pith-number/6DUBUA4WDPQEGG47BJ7YOROKVJ/graph.json","fetch_events":"https://pith.science/api/pith-number/6DUBUA4WDPQEGG47BJ7YOROKVJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ/action/storage_attestation","attest_author":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ/action/author_attestation","sign_citation":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ/action/citation_signature","submit_replication":"https://pith.science/pith/6DUBUA4WDPQEGG47BJ7YOROKVJ/action/replication_record"}},"created_at":"2026-07-05T01:27:32.095335+00:00","updated_at":"2026-07-05T01:27:32.095335+00:00"}