{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PB26J7ZZWDOJ7SIUM3JYOJ56DR","short_pith_number":"pith:PB26J7ZZ","schema_version":"1.0","canonical_sha256":"7875e4ff39b0dc9fc91466d38727be1c4733bc8fcac3fa1d13bfbe4b821eff7f","source":{"kind":"arxiv","id":"2303.02011","version":4},"attestation_state":"computed","paper":{"title":"Diagnosing Model Performance Under Distribution Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Hongseok Namkoong, Steve Yadlowsky, Tiffany Tianhui Cai","submitted_at":"2023-03-03T15:27:16Z","abstract_excerpt":"Prediction models can perform poorly when deployed to target distributions different from the training distribution. To understand these operational failure modes, we develop a method, called DIstribution Shift DEcomposition (DISDE), to attribute a drop in performance to different types of distribution shifts. Our approach decomposes the performance drop into terms for 1) an increase in harder but frequently seen examples from training, 2) changes in the relationship between features and outcomes, and 3) poor performance on examples infrequent or unseen during training. These terms are defined"},"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":"2303.02011","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2023-03-03T15:27:16Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d7b6594c7e0fa59efe646fda7a4a9e99fb2c0e28cf3229017a24553ddbd5567c","abstract_canon_sha256":"cf852b2c37de16c82ccb1bc13b2d965eee3eba9b82a709c9d5ee3279ed1cbf32"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:29:20.138405Z","signature_b64":"S/uz7rFDNMplACRTIlJ1czieQs3PQcBEYVGaIIgR5NOhPoKoXGzZmin3rscSOeHIhNoVaHa0fKIWZrQ/9+U+DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7875e4ff39b0dc9fc91466d38727be1c4733bc8fcac3fa1d13bfbe4b821eff7f","last_reissued_at":"2026-07-05T06:29:20.137955Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:29:20.137955Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diagnosing Model Performance Under Distribution Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Hongseok Namkoong, Steve Yadlowsky, Tiffany Tianhui Cai","submitted_at":"2023-03-03T15:27:16Z","abstract_excerpt":"Prediction models can perform poorly when deployed to target distributions different from the training distribution. To understand these operational failure modes, we develop a method, called DIstribution Shift DEcomposition (DISDE), to attribute a drop in performance to different types of distribution shifts. Our approach decomposes the performance drop into terms for 1) an increase in harder but frequently seen examples from training, 2) changes in the relationship between features and outcomes, and 3) poor performance on examples infrequent or unseen during training. These terms are defined"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.02011","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/2303.02011/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":"2303.02011","created_at":"2026-07-05T06:29:20.138010+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.02011v4","created_at":"2026-07-05T06:29:20.138010+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.02011","created_at":"2026-07-05T06:29:20.138010+00:00"},{"alias_kind":"pith_short_12","alias_value":"PB26J7ZZWDOJ","created_at":"2026-07-05T06:29:20.138010+00:00"},{"alias_kind":"pith_short_16","alias_value":"PB26J7ZZWDOJ7SIU","created_at":"2026-07-05T06:29:20.138010+00:00"},{"alias_kind":"pith_short_8","alias_value":"PB26J7ZZ","created_at":"2026-07-05T06:29:20.138010+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.11200","citing_title":"ShapShift: Explaining Model Prediction Shifts with Subgroup Conditional Shapley Values","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR","json":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR.json","graph_json":"https://pith.science/api/pith-number/PB26J7ZZWDOJ7SIUM3JYOJ56DR/graph.json","events_json":"https://pith.science/api/pith-number/PB26J7ZZWDOJ7SIUM3JYOJ56DR/events.json","paper":"https://pith.science/paper/PB26J7ZZ"},"agent_actions":{"view_html":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR","download_json":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR.json","view_paper":"https://pith.science/paper/PB26J7ZZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.02011&json=true","fetch_graph":"https://pith.science/api/pith-number/PB26J7ZZWDOJ7SIUM3JYOJ56DR/graph.json","fetch_events":"https://pith.science/api/pith-number/PB26J7ZZWDOJ7SIUM3JYOJ56DR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR/action/storage_attestation","attest_author":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR/action/author_attestation","sign_citation":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR/action/citation_signature","submit_replication":"https://pith.science/pith/PB26J7ZZWDOJ7SIUM3JYOJ56DR/action/replication_record"}},"created_at":"2026-07-05T06:29:20.138010+00:00","updated_at":"2026-07-05T06:29:20.138010+00:00"}