{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HE4XXCVFN4EPLRIR4EMD4YYTAJ","short_pith_number":"pith:HE4XXCVF","schema_version":"1.0","canonical_sha256":"39397b8aa56f08f5c511e1183e6313025f3bee5bf9e1a79e1835a4702dd1f164","source":{"kind":"arxiv","id":"2503.18334","version":2},"attestation_state":"computed","paper":{"title":"Mitigating Cache Noise in Test-Time Adaptation for Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Can Zhang, Haotian Zhai, Ruirui Li, Tianming Sha, Xinyu Chen","submitted_at":"2025-03-24T04:32:35Z","abstract_excerpt":"Test-time adaptation (TTA) of visual language models has recently attracted significant attention as a solution to the performance degradation caused by distribution shifts in downstream tasks. However, existing cache-based TTA methods have certain limitations. They mainly rely on the accuracy of cached feature labels, and the presence of noisy pseudo-labels can cause these features to deviate from their true distribution. This makes cache retrieval methods based on similarity matching highly sensitive to outliers or extreme samples. Moreover, current methods lack effective mechanisms to model"},"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":"2503.18334","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-24T04:32:35Z","cross_cats_sorted":[],"title_canon_sha256":"ccc2b1402f3415d075282c41bdcac811f93f9fd9cbedf0bbe6ff70cdf5ece09f","abstract_canon_sha256":"607ea3c3dc2b7407c9678ec3191ad618e9f6a989165c2d17a77871608417131d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:41:53.971170Z","signature_b64":"OPQVinX+Q81avCn5887939ULy8PeV4GxTzg7dB5Tv4LUIYevSo8Uiqg8ifaxu2FapY1CsYpmmJPyjvrqiXNBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39397b8aa56f08f5c511e1183e6313025f3bee5bf9e1a79e1835a4702dd1f164","last_reissued_at":"2026-07-05T10:41:53.970683Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:41:53.970683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mitigating Cache Noise in Test-Time Adaptation for Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Can Zhang, Haotian Zhai, Ruirui Li, Tianming Sha, Xinyu Chen","submitted_at":"2025-03-24T04:32:35Z","abstract_excerpt":"Test-time adaptation (TTA) of visual language models has recently attracted significant attention as a solution to the performance degradation caused by distribution shifts in downstream tasks. However, existing cache-based TTA methods have certain limitations. They mainly rely on the accuracy of cached feature labels, and the presence of noisy pseudo-labels can cause these features to deviate from their true distribution. This makes cache retrieval methods based on similarity matching highly sensitive to outliers or extreme samples. Moreover, current methods lack effective mechanisms to model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.18334","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/2503.18334/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":"2503.18334","created_at":"2026-07-05T10:41:53.970741+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.18334v2","created_at":"2026-07-05T10:41:53.970741+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.18334","created_at":"2026-07-05T10:41:53.970741+00:00"},{"alias_kind":"pith_short_12","alias_value":"HE4XXCVFN4EP","created_at":"2026-07-05T10:41:53.970741+00:00"},{"alias_kind":"pith_short_16","alias_value":"HE4XXCVFN4EPLRIR","created_at":"2026-07-05T10:41:53.970741+00:00"},{"alias_kind":"pith_short_8","alias_value":"HE4XXCVF","created_at":"2026-07-05T10:41:53.970741+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.08887","citing_title":"FAST-CAD: A Fairness-Aware Framework for Non-Contact Stroke Diagnosis","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ","json":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ.json","graph_json":"https://pith.science/api/pith-number/HE4XXCVFN4EPLRIR4EMD4YYTAJ/graph.json","events_json":"https://pith.science/api/pith-number/HE4XXCVFN4EPLRIR4EMD4YYTAJ/events.json","paper":"https://pith.science/paper/HE4XXCVF"},"agent_actions":{"view_html":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ","download_json":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ.json","view_paper":"https://pith.science/paper/HE4XXCVF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.18334&json=true","fetch_graph":"https://pith.science/api/pith-number/HE4XXCVFN4EPLRIR4EMD4YYTAJ/graph.json","fetch_events":"https://pith.science/api/pith-number/HE4XXCVFN4EPLRIR4EMD4YYTAJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ/action/storage_attestation","attest_author":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ/action/author_attestation","sign_citation":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ/action/citation_signature","submit_replication":"https://pith.science/pith/HE4XXCVFN4EPLRIR4EMD4YYTAJ/action/replication_record"}},"created_at":"2026-07-05T10:41:53.970741+00:00","updated_at":"2026-07-05T10:41:53.970741+00:00"}