{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PE42D7Y7GBFTSTYGPTI5LDOCTS","short_pith_number":"pith:PE42D7Y7","schema_version":"1.0","canonical_sha256":"7939a1ff1f304b394f067cd1d58dc29cb453ca6e6e71d88599df1461db0fd2a2","source":{"kind":"arxiv","id":"2301.13018","version":1},"attestation_state":"computed","paper":{"title":"DELTA: degradation-free fully test-time adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Bowen Zhao, Chen Chen, Shu-Tao Xia","submitted_at":"2023-01-30T15:54:00Z","abstract_excerpt":"Fully test-time adaptation aims at adapting a pre-trained model to the test stream during real-time inference, which is urgently required when the test distribution differs from the training distribution. Several efforts have been devoted to improving adaptation performance. However, we find that two unfavorable defects are concealed in the prevalent adaptation methodologies like test-time batch normalization (BN) and self-learning. First, we reveal that the normalization statistics in test-time BN are completely affected by the currently received test samples, resulting in inaccurate estimate"},"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":"2301.13018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-30T15:54:00Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"56269146f4cec830c3b719031c108ce5ccc38004b7102ffd715a4c2a684f0dae","abstract_canon_sha256":"6fe2c6f50ebab46b2e9a7bd4f0c38b10ad2bba0592cbd383349323157f03e31a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:37:00.135116Z","signature_b64":"M1zb0zHddKt/hKrfO6kkm9sIfc6p+wt0v5ksmaT6DBWz6X4uCPzdtcdU7HkMrcrnmUp9LATdItZH+zYFEARRBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7939a1ff1f304b394f067cd1d58dc29cb453ca6e6e71d88599df1461db0fd2a2","last_reissued_at":"2026-07-05T05:37:00.134631Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:37:00.134631Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DELTA: degradation-free fully test-time adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Bowen Zhao, Chen Chen, Shu-Tao Xia","submitted_at":"2023-01-30T15:54:00Z","abstract_excerpt":"Fully test-time adaptation aims at adapting a pre-trained model to the test stream during real-time inference, which is urgently required when the test distribution differs from the training distribution. Several efforts have been devoted to improving adaptation performance. However, we find that two unfavorable defects are concealed in the prevalent adaptation methodologies like test-time batch normalization (BN) and self-learning. First, we reveal that the normalization statistics in test-time BN are completely affected by the currently received test samples, resulting in inaccurate estimate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.13018","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/2301.13018/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":"2301.13018","created_at":"2026-07-05T05:37:00.134698+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.13018v1","created_at":"2026-07-05T05:37:00.134698+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.13018","created_at":"2026-07-05T05:37:00.134698+00:00"},{"alias_kind":"pith_short_12","alias_value":"PE42D7Y7GBFT","created_at":"2026-07-05T05:37:00.134698+00:00"},{"alias_kind":"pith_short_16","alias_value":"PE42D7Y7GBFTSTYG","created_at":"2026-07-05T05:37:00.134698+00:00"},{"alias_kind":"pith_short_8","alias_value":"PE42D7Y7","created_at":"2026-07-05T05:37:00.134698+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25758","citing_title":"Dual Distribution Estimation for Zero-shot Noisy Test-Time Adaptation with VLMs","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25651","citing_title":"Hierarchical Consistency Learning for Test-time Adaptation in Camouflage Perception","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2508.15452","citing_title":"DoSReMC: Domain Shift Resilient Mammography Classification using Batch Normalization Adaptation","ref_index":70,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS","json":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS.json","graph_json":"https://pith.science/api/pith-number/PE42D7Y7GBFTSTYGPTI5LDOCTS/graph.json","events_json":"https://pith.science/api/pith-number/PE42D7Y7GBFTSTYGPTI5LDOCTS/events.json","paper":"https://pith.science/paper/PE42D7Y7"},"agent_actions":{"view_html":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS","download_json":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS.json","view_paper":"https://pith.science/paper/PE42D7Y7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.13018&json=true","fetch_graph":"https://pith.science/api/pith-number/PE42D7Y7GBFTSTYGPTI5LDOCTS/graph.json","fetch_events":"https://pith.science/api/pith-number/PE42D7Y7GBFTSTYGPTI5LDOCTS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS/action/storage_attestation","attest_author":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS/action/author_attestation","sign_citation":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS/action/citation_signature","submit_replication":"https://pith.science/pith/PE42D7Y7GBFTSTYGPTI5LDOCTS/action/replication_record"}},"created_at":"2026-07-05T05:37:00.134698+00:00","updated_at":"2026-07-05T05:37:00.134698+00:00"}