{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:45HSVJAQAS6GB3JSHNXZC2BVPS","short_pith_number":"pith:45HSVJAQ","schema_version":"1.0","canonical_sha256":"e74f2aa41004bc60ed323b6f9168357ca70120ed343cc0a76ad9044634ee5b44","source":{"kind":"arxiv","id":"2209.00613","version":4},"attestation_state":"computed","paper":{"title":"ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Damien Teney, Ehsan Abbasnejad, Seong Joon Oh, Yong Lin","submitted_at":"2022-09-01T17:27:25Z","abstract_excerpt":"Several studies have compared the in-distribution (ID) and out-of-distribution (OOD) performance of models in computer vision and NLP. They report a frequent positive correlation and some surprisingly never even observe an inverse correlation indicative of a necessary trade-off. The possibility of inverse patterns is important to determine whether ID performance can serve as a proxy for OOD generalization capabilities.\n  This paper shows with multiple datasets that inverse correlations between ID and OOD performance do happen in real-world data - not only in theoretical worst-case settings. We"},"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":"2209.00613","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-09-01T17:27:25Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"aafe372db5781c6ac20ba44204a923334584eb1aabdd1bec162848c75dc99558","abstract_canon_sha256":"05e1beaa6f4d504c69b47dadbbfaf3ebbfa0f3cbee77b37be403e01e16c4dd46"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:11:42.584552Z","signature_b64":"vrc1m7PdWmMrYY3Z4K276pIBeDSDSc6+4vYy+7uhs+7EYdS2EKwAkefqfgwz5yF+qKam6BwBa73F/fg2coovCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e74f2aa41004bc60ed323b6f9168357ca70120ed343cc0a76ad9044634ee5b44","last_reissued_at":"2026-07-05T06:11:42.584205Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:11:42.584205Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Damien Teney, Ehsan Abbasnejad, Seong Joon Oh, Yong Lin","submitted_at":"2022-09-01T17:27:25Z","abstract_excerpt":"Several studies have compared the in-distribution (ID) and out-of-distribution (OOD) performance of models in computer vision and NLP. They report a frequent positive correlation and some surprisingly never even observe an inverse correlation indicative of a necessary trade-off. The possibility of inverse patterns is important to determine whether ID performance can serve as a proxy for OOD generalization capabilities.\n  This paper shows with multiple datasets that inverse correlations between ID and OOD performance do happen in real-world data - not only in theoretical worst-case settings. We"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.00613","kind":"arxiv","version":4},"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/2209.00613/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":"2209.00613","created_at":"2026-07-05T06:11:42.584260+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.00613v4","created_at":"2026-07-05T06:11:42.584260+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.00613","created_at":"2026-07-05T06:11:42.584260+00:00"},{"alias_kind":"pith_short_12","alias_value":"45HSVJAQAS6G","created_at":"2026-07-05T06:11:42.584260+00:00"},{"alias_kind":"pith_short_16","alias_value":"45HSVJAQAS6GB3JS","created_at":"2026-07-05T06:11:42.584260+00:00"},{"alias_kind":"pith_short_8","alias_value":"45HSVJAQ","created_at":"2026-07-05T06:11:42.584260+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.12226","citing_title":"Learning Causality for Modern Machine Learning","ref_index":62,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS","json":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS.json","graph_json":"https://pith.science/api/pith-number/45HSVJAQAS6GB3JSHNXZC2BVPS/graph.json","events_json":"https://pith.science/api/pith-number/45HSVJAQAS6GB3JSHNXZC2BVPS/events.json","paper":"https://pith.science/paper/45HSVJAQ"},"agent_actions":{"view_html":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS","download_json":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS.json","view_paper":"https://pith.science/paper/45HSVJAQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.00613&json=true","fetch_graph":"https://pith.science/api/pith-number/45HSVJAQAS6GB3JSHNXZC2BVPS/graph.json","fetch_events":"https://pith.science/api/pith-number/45HSVJAQAS6GB3JSHNXZC2BVPS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS/action/storage_attestation","attest_author":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS/action/author_attestation","sign_citation":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS/action/citation_signature","submit_replication":"https://pith.science/pith/45HSVJAQAS6GB3JSHNXZC2BVPS/action/replication_record"}},"created_at":"2026-07-05T06:11:42.584260+00:00","updated_at":"2026-07-05T06:11:42.584260+00:00"}