{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:O3M6242UM2BUJUA2FQVFY5CIVR","short_pith_number":"pith:O3M6242U","schema_version":"1.0","canonical_sha256":"76d9ed7354668344d01a2c2a5c7448ac61f2bc0b584c720605b439a38d03ab76","source":{"kind":"arxiv","id":"2112.04094","version":1},"attestation_state":"computed","paper":{"title":"The Effect of Model Size on Worst-Group Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alan Pham, Dhruba Ghosh, Eunice Chan, Jacob Steinhardt, Joseph E. Gonzalez, Ruiqi Zhong, Vikranth Srivatsa, Yaodong Yu, Yaoqing Yang","submitted_at":"2021-12-08T03:45:47Z","abstract_excerpt":"Overparameterization is shown to result in poor test accuracy on rare subgroups under a variety of settings where subgroup information is known. To gain a more complete picture, we consider the case where subgroup information is unknown. We investigate the effect of model size on worst-group generalization under empirical risk minimization (ERM) across a wide range of settings, varying: 1) architectures (ResNet, VGG, or BERT), 2) domains (vision or natural language processing), 3) model size (width or depth), and 4) initialization (with pre-trained or random weights). Our systematic evaluation"},"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":"2112.04094","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-08T03:45:47Z","cross_cats_sorted":[],"title_canon_sha256":"7ecece3bdb2670f29d595d6dd1c50f5e9e9b9bdeffd98865019c075b05a4b6be","abstract_canon_sha256":"4d46c8724b8066bcf542291ab22e55046fdfc954915000385f5ea287433c1cb5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:39:05.948715Z","signature_b64":"N99/H34HUZGh+BedbSoFE8aKhCPJ2XY3Jqsgx1TA3djXYZRUWWf+Fm6nBsWJBNDjkw2hOeOsC8Rp/9TXKF35Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"76d9ed7354668344d01a2c2a5c7448ac61f2bc0b584c720605b439a38d03ab76","last_reissued_at":"2026-07-05T03:39:05.948369Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:39:05.948369Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Effect of Model Size on Worst-Group Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alan Pham, Dhruba Ghosh, Eunice Chan, Jacob Steinhardt, Joseph E. Gonzalez, Ruiqi Zhong, Vikranth Srivatsa, Yaodong Yu, Yaoqing Yang","submitted_at":"2021-12-08T03:45:47Z","abstract_excerpt":"Overparameterization is shown to result in poor test accuracy on rare subgroups under a variety of settings where subgroup information is known. To gain a more complete picture, we consider the case where subgroup information is unknown. We investigate the effect of model size on worst-group generalization under empirical risk minimization (ERM) across a wide range of settings, varying: 1) architectures (ResNet, VGG, or BERT), 2) domains (vision or natural language processing), 3) model size (width or depth), and 4) initialization (with pre-trained or random weights). Our systematic evaluation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.04094","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/2112.04094/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":"2112.04094","created_at":"2026-07-05T03:39:05.948423+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.04094v1","created_at":"2026-07-05T03:39:05.948423+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.04094","created_at":"2026-07-05T03:39:05.948423+00:00"},{"alias_kind":"pith_short_12","alias_value":"O3M6242UM2BU","created_at":"2026-07-05T03:39:05.948423+00:00"},{"alias_kind":"pith_short_16","alias_value":"O3M6242UM2BUJUA2","created_at":"2026-07-05T03:39:05.948423+00:00"},{"alias_kind":"pith_short_8","alias_value":"O3M6242U","created_at":"2026-07-05T03:39:05.948423+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR","json":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR.json","graph_json":"https://pith.science/api/pith-number/O3M6242UM2BUJUA2FQVFY5CIVR/graph.json","events_json":"https://pith.science/api/pith-number/O3M6242UM2BUJUA2FQVFY5CIVR/events.json","paper":"https://pith.science/paper/O3M6242U"},"agent_actions":{"view_html":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR","download_json":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR.json","view_paper":"https://pith.science/paper/O3M6242U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.04094&json=true","fetch_graph":"https://pith.science/api/pith-number/O3M6242UM2BUJUA2FQVFY5CIVR/graph.json","fetch_events":"https://pith.science/api/pith-number/O3M6242UM2BUJUA2FQVFY5CIVR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR/action/storage_attestation","attest_author":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR/action/author_attestation","sign_citation":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR/action/citation_signature","submit_replication":"https://pith.science/pith/O3M6242UM2BUJUA2FQVFY5CIVR/action/replication_record"}},"created_at":"2026-07-05T03:39:05.948423+00:00","updated_at":"2026-07-05T03:39:05.948423+00:00"}