{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ES4A666XLINP7ATRW63F3QHI3U","short_pith_number":"pith:ES4A666X","schema_version":"1.0","canonical_sha256":"24b80f7bd75a1aff8271b7b65dc0e8dd22b8a1a84ba03f55c8489cb518f50bb1","source":{"kind":"arxiv","id":"2401.08328","version":2},"attestation_state":"computed","paper":{"title":"Un-Mixing Test-Time Normalization Statistics: Combatting Label Temporal Correlation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Behzad Bozorgtabar, Devavrat Tomar, Guillaume Vray, Jean-Philippe Thiran","submitted_at":"2024-01-16T12:48:52Z","abstract_excerpt":"Recent test-time adaptation methods heavily rely on nuanced adjustments of batch normalization (BN) parameters. However, one critical assumption often goes overlooked: that of independently and identically distributed (i.i.d.) test batches with respect to unknown labels. This oversight leads to skewed BN statistics and undermines the reliability of the model under non-i.i.d. scenarios. To tackle this challenge, this paper presents a novel method termed 'Un-Mixing Test-Time Normalization Statistics' (UnMix-TNS). Our method re-calibrates the statistics for each instance within a test batch by mi"},"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":"2401.08328","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-16T12:48:52Z","cross_cats_sorted":[],"title_canon_sha256":"743f48c5fc21efc2a4bd39f535e945b7d5d62a99579f1f4952016083e7e15d7a","abstract_canon_sha256":"2a2fa48327274599a606bb120560b08ec74fbc1ca0e941f1bdc88ea56ebb1f65"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:56:02.636043Z","signature_b64":"EmkmPCqz0sRB5SFEJKCrvWdjryyB1nyHI/SJlD5MnFHZqMUkgKXuPsHIOwxu1Ts6qGLyyOG3/+WI24vAiAhnCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24b80f7bd75a1aff8271b7b65dc0e8dd22b8a1a84ba03f55c8489cb518f50bb1","last_reissued_at":"2026-07-05T07:56:02.635558Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:56:02.635558Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Un-Mixing Test-Time Normalization Statistics: Combatting Label Temporal Correlation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Behzad Bozorgtabar, Devavrat Tomar, Guillaume Vray, Jean-Philippe Thiran","submitted_at":"2024-01-16T12:48:52Z","abstract_excerpt":"Recent test-time adaptation methods heavily rely on nuanced adjustments of batch normalization (BN) parameters. However, one critical assumption often goes overlooked: that of independently and identically distributed (i.i.d.) test batches with respect to unknown labels. This oversight leads to skewed BN statistics and undermines the reliability of the model under non-i.i.d. scenarios. To tackle this challenge, this paper presents a novel method termed 'Un-Mixing Test-Time Normalization Statistics' (UnMix-TNS). Our method re-calibrates the statistics for each instance within a test batch by mi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.08328","kind":"arxiv","version":2},"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/2401.08328/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":"2401.08328","created_at":"2026-07-05T07:56:02.635616+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.08328v2","created_at":"2026-07-05T07:56:02.635616+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.08328","created_at":"2026-07-05T07:56:02.635616+00:00"},{"alias_kind":"pith_short_12","alias_value":"ES4A666XLINP","created_at":"2026-07-05T07:56:02.635616+00:00"},{"alias_kind":"pith_short_16","alias_value":"ES4A666XLINP7ATR","created_at":"2026-07-05T07:56:02.635616+00:00"},{"alias_kind":"pith_short_8","alias_value":"ES4A666X","created_at":"2026-07-05T07:56:02.635616+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.18751","citing_title":"Stabilizing Open-Set Test-Time Adaptation via Primary-Auxiliary Filtering and Knowledge-Integrated Prediction","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U","json":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U.json","graph_json":"https://pith.science/api/pith-number/ES4A666XLINP7ATRW63F3QHI3U/graph.json","events_json":"https://pith.science/api/pith-number/ES4A666XLINP7ATRW63F3QHI3U/events.json","paper":"https://pith.science/paper/ES4A666X"},"agent_actions":{"view_html":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U","download_json":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U.json","view_paper":"https://pith.science/paper/ES4A666X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.08328&json=true","fetch_graph":"https://pith.science/api/pith-number/ES4A666XLINP7ATRW63F3QHI3U/graph.json","fetch_events":"https://pith.science/api/pith-number/ES4A666XLINP7ATRW63F3QHI3U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U/action/storage_attestation","attest_author":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U/action/author_attestation","sign_citation":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U/action/citation_signature","submit_replication":"https://pith.science/pith/ES4A666XLINP7ATRW63F3QHI3U/action/replication_record"}},"created_at":"2026-07-05T07:56:02.635616+00:00","updated_at":"2026-07-05T07:56:02.635616+00:00"}