{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:MGG3QSN42KSOMI3EINSWJKKIVU","short_pith_number":"pith:MGG3QSN4","schema_version":"1.0","canonical_sha256":"618db849bcd2a4e62364436564a948ad1495906c43e7dca734512fbc23462e83","source":{"kind":"arxiv","id":"2110.04065","version":1},"attestation_state":"computed","paper":{"title":"Test-time Batch Statistics Calibration for Covariate Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Fuming You, Jingjing Li, Zhou Zhao","submitted_at":"2021-10-06T08:45:03Z","abstract_excerpt":"Deep neural networks have a clear degradation when applying to the unseen environment due to the covariate shift. Conventional approaches like domain adaptation requires the pre-collected target data for iterative training, which is impractical in real-world applications. In this paper, we propose to adapt the deep models to the novel environment during inference. An previous solution is test time normalization, which substitutes the source statistics in BN layers with the target batch statistics. However, we show that test time normalization may potentially deteriorate the discriminative stru"},"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":"2110.04065","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-06T08:45:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d4f67c683686904bfb29c8ff209b4e3612172699663061a64e0c91437e03eed9","abstract_canon_sha256":"2e5e99d1fcccb16fbb45396d6d6bb3354c2c8f00e9356d8f0de8cadaeeaf14ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:21:00.563745Z","signature_b64":"bWkbjxB/sWiHYyvAaFnJKJO9Z0XsnmRQAseUDE7nGi/AmEaLT50UyQw8MvK2F527EoiWmG6c4hm5Hf4heMWGDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"618db849bcd2a4e62364436564a948ad1495906c43e7dca734512fbc23462e83","last_reissued_at":"2026-07-05T03:21:00.563338Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:21:00.563338Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Test-time Batch Statistics Calibration for Covariate Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Fuming You, Jingjing Li, Zhou Zhao","submitted_at":"2021-10-06T08:45:03Z","abstract_excerpt":"Deep neural networks have a clear degradation when applying to the unseen environment due to the covariate shift. Conventional approaches like domain adaptation requires the pre-collected target data for iterative training, which is impractical in real-world applications. In this paper, we propose to adapt the deep models to the novel environment during inference. An previous solution is test time normalization, which substitutes the source statistics in BN layers with the target batch statistics. However, we show that test time normalization may potentially deteriorate the discriminative stru"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.04065","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/2110.04065/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":"2110.04065","created_at":"2026-07-05T03:21:00.563400+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.04065v1","created_at":"2026-07-05T03:21:00.563400+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.04065","created_at":"2026-07-05T03:21:00.563400+00:00"},{"alias_kind":"pith_short_12","alias_value":"MGG3QSN42KSO","created_at":"2026-07-05T03:21:00.563400+00:00"},{"alias_kind":"pith_short_16","alias_value":"MGG3QSN42KSOMI3E","created_at":"2026-07-05T03:21:00.563400+00:00"},{"alias_kind":"pith_short_8","alias_value":"MGG3QSN4","created_at":"2026-07-05T03:21:00.563400+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18867","citing_title":"EVA-0: Test-Time Model Evolution with Only Two Forward Passes per Sample","ref_index":47,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU","json":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU.json","graph_json":"https://pith.science/api/pith-number/MGG3QSN42KSOMI3EINSWJKKIVU/graph.json","events_json":"https://pith.science/api/pith-number/MGG3QSN42KSOMI3EINSWJKKIVU/events.json","paper":"https://pith.science/paper/MGG3QSN4"},"agent_actions":{"view_html":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU","download_json":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU.json","view_paper":"https://pith.science/paper/MGG3QSN4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.04065&json=true","fetch_graph":"https://pith.science/api/pith-number/MGG3QSN42KSOMI3EINSWJKKIVU/graph.json","fetch_events":"https://pith.science/api/pith-number/MGG3QSN42KSOMI3EINSWJKKIVU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU/action/storage_attestation","attest_author":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU/action/author_attestation","sign_citation":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU/action/citation_signature","submit_replication":"https://pith.science/pith/MGG3QSN42KSOMI3EINSWJKKIVU/action/replication_record"}},"created_at":"2026-07-05T03:21:00.563400+00:00","updated_at":"2026-07-05T03:21:00.563400+00:00"}