{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4PHDRBRV2Y4V6M6GY4ME7EYZDX","short_pith_number":"pith:4PHDRBRV","canonical_record":{"source":{"id":"2508.05898","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-07T23:11:33Z","cross_cats_sorted":[],"title_canon_sha256":"6a02bdab91d816bb977036d5c3680997690063b575dcc0d205fb926c867272ce","abstract_canon_sha256":"2d6399214ee3807035b4f89530677dbc8d8081c0b05734c5e956d6c6554e913f"},"schema_version":"1.0"},"canonical_sha256":"e3ce388635d6395f33c6c7184f93191dc61408918e9da1444e5f4bb1e12612dd","source":{"kind":"arxiv","id":"2508.05898","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.05898","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"arxiv_version","alias_value":"2508.05898v1","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05898","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"pith_short_12","alias_value":"4PHDRBRV2Y4V","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"pith_short_16","alias_value":"4PHDRBRV2Y4V6M6G","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"pith_short_8","alias_value":"4PHDRBRV","created_at":"2026-07-05T11:50:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4PHDRBRV2Y4V6M6GY4ME7EYZDX","target":"record","payload":{"canonical_record":{"source":{"id":"2508.05898","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-07T23:11:33Z","cross_cats_sorted":[],"title_canon_sha256":"6a02bdab91d816bb977036d5c3680997690063b575dcc0d205fb926c867272ce","abstract_canon_sha256":"2d6399214ee3807035b4f89530677dbc8d8081c0b05734c5e956d6c6554e913f"},"schema_version":"1.0"},"canonical_sha256":"e3ce388635d6395f33c6c7184f93191dc61408918e9da1444e5f4bb1e12612dd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:45.121018Z","signature_b64":"SGyzwMiP7PmqPcOwwviFRATvTXDz01866WnmTXpWcsdCdT90VtcQT/bcngIAE9NMYMTTyGzQosP/IA1Pr5qVBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3ce388635d6395f33c6c7184f93191dc61408918e9da1444e5f4bb1e12612dd","last_reissued_at":"2026-07-05T11:50:45.120570Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:45.120570Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.05898","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:50:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p3U4o0nqonZCK1b3geA6hJ91zyu46Q7QzyE3M07GudF9T+KfUCaEUKJMrVdh7GVPmumfcX0Vnc4QjuxN3aMeDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T12:27:05.918514Z"},"content_sha256":"25e6d07393716f4fb55a92c99466cdc9f116eb2653c95033e6e7250582796229","schema_version":"1.0","event_id":"sha256:25e6d07393716f4fb55a92c99466cdc9f116eb2653c95033e6e7250582796229"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4PHDRBRV2Y4V6M6GY4ME7EYZDX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ETTA: Efficient Test-Time Adaptation for Vision-Language Models through Dynamic Embedding Updates","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aijun An, Ali Cheraghian, Hamidreza Dastmalchi","submitted_at":"2025-08-07T23:11:33Z","abstract_excerpt":"Pretrained vision-language models (VLMs) like CLIP show strong zero-shot performance but struggle with generalization under distribution shifts. Test-Time Adaptation (TTA) addresses this by adapting VLMs to unlabeled test data in new domains. While some TTA methods rely on prompt-tuning, training-free cache-based approaches are preferred for efficiency. However, current cache-based TTA models store only a limited set of high-confidence samples, restricting the decision boundary to these samples and ignoring the influence of other incoming test data. To address this, we propose Efficient Test-T"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05898","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/2508.05898/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:50:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yoeXmZO0vodEorxIANFTCMAYh9SMxIpcqKhnzwzlH5r0zmptGQVEpU0r+XLWKpUZ5J4QLGv9XsK6YWaSWq9RBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T12:27:05.919437Z"},"content_sha256":"0c92c241b7c7d7d2212fe42b6cb19d2759519d2473d452dfdf1914aaa30334d4","schema_version":"1.0","event_id":"sha256:0c92c241b7c7d7d2212fe42b6cb19d2759519d2473d452dfdf1914aaa30334d4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX/bundle.json","state_url":"https://pith.science/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-14T12:27:05Z","links":{"resolver":"https://pith.science/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX","bundle":"https://pith.science/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX/bundle.json","state":"https://pith.science/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4PHDRBRV2Y4V6M6GY4ME7EYZDX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4PHDRBRV2Y4V6M6GY4ME7EYZDX","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2d6399214ee3807035b4f89530677dbc8d8081c0b05734c5e956d6c6554e913f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-07T23:11:33Z","title_canon_sha256":"6a02bdab91d816bb977036d5c3680997690063b575dcc0d205fb926c867272ce"},"schema_version":"1.0","source":{"id":"2508.05898","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.05898","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"arxiv_version","alias_value":"2508.05898v1","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05898","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"pith_short_12","alias_value":"4PHDRBRV2Y4V","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"pith_short_16","alias_value":"4PHDRBRV2Y4V6M6G","created_at":"2026-07-05T11:50:45Z"},{"alias_kind":"pith_short_8","alias_value":"4PHDRBRV","created_at":"2026-07-05T11:50:45Z"}],"graph_snapshots":[{"event_id":"sha256:0c92c241b7c7d7d2212fe42b6cb19d2759519d2473d452dfdf1914aaa30334d4","target":"graph","created_at":"2026-07-05T11:50:45Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2508.05898/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pretrained vision-language models (VLMs) like CLIP show strong zero-shot performance but struggle with generalization under distribution shifts. Test-Time Adaptation (TTA) addresses this by adapting VLMs to unlabeled test data in new domains. While some TTA methods rely on prompt-tuning, training-free cache-based approaches are preferred for efficiency. However, current cache-based TTA models store only a limited set of high-confidence samples, restricting the decision boundary to these samples and ignoring the influence of other incoming test data. To address this, we propose Efficient Test-T","authors_text":"Aijun An, Ali Cheraghian, Hamidreza Dastmalchi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-07T23:11:33Z","title":"ETTA: Efficient Test-Time Adaptation for Vision-Language Models through Dynamic Embedding Updates"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05898","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:25e6d07393716f4fb55a92c99466cdc9f116eb2653c95033e6e7250582796229","target":"record","created_at":"2026-07-05T11:50:45Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2d6399214ee3807035b4f89530677dbc8d8081c0b05734c5e956d6c6554e913f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-07T23:11:33Z","title_canon_sha256":"6a02bdab91d816bb977036d5c3680997690063b575dcc0d205fb926c867272ce"},"schema_version":"1.0","source":{"id":"2508.05898","kind":"arxiv","version":1}},"canonical_sha256":"e3ce388635d6395f33c6c7184f93191dc61408918e9da1444e5f4bb1e12612dd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3ce388635d6395f33c6c7184f93191dc61408918e9da1444e5f4bb1e12612dd","first_computed_at":"2026-07-05T11:50:45.120570Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:50:45.120570Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SGyzwMiP7PmqPcOwwviFRATvTXDz01866WnmTXpWcsdCdT90VtcQT/bcngIAE9NMYMTTyGzQosP/IA1Pr5qVBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:50:45.121018Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.05898","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:25e6d07393716f4fb55a92c99466cdc9f116eb2653c95033e6e7250582796229","sha256:0c92c241b7c7d7d2212fe42b6cb19d2759519d2473d452dfdf1914aaa30334d4"],"state_sha256":"6a966dbc54a1cbe83172d16991e2f2e601c2f0590db29d00232c771c11a10052"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h0ONPWU7gsFG/wv2ZGKHYl4l0NzQM5zbnU5rPOwfEZSAdt4+mGxQZQSwau51WtIgLdhrd4hTThE/HSrcOJOQBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T12:27:05.937252Z","bundle_sha256":"085fd925802de19f9f496986e5848a3ef4625773aaebb154b761f6049d0ea9ca"}}