{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:NI2XYUP3KRZ3YAC7VQDA6NIQ6Z","short_pith_number":"pith:NI2XYUP3","canonical_record":{"source":{"id":"2209.07511","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-09-15T17:55:11Z","cross_cats_sorted":[],"title_canon_sha256":"25e990427c0c07f3a48b5e8af98a51d54688afab2a93a54832617f0c29ab4e8a","abstract_canon_sha256":"86afbada7c3cf0ff22e46ffd91b405ea4ba9cb155164ac3b0ab461a466058d04"},"schema_version":"1.0"},"canonical_sha256":"6a357c51fb5473bc005fac060f3510f6670afdf27fe52f895f30b84ed38b8943","source":{"kind":"arxiv","id":"2209.07511","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2209.07511","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"arxiv_version","alias_value":"2209.07511v1","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.07511","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"pith_short_12","alias_value":"NI2XYUP3KRZ3","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"pith_short_16","alias_value":"NI2XYUP3KRZ3YAC7","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"pith_short_8","alias_value":"NI2XYUP3","created_at":"2026-07-05T04:57:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:NI2XYUP3KRZ3YAC7VQDA6NIQ6Z","target":"record","payload":{"canonical_record":{"source":{"id":"2209.07511","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-09-15T17:55:11Z","cross_cats_sorted":[],"title_canon_sha256":"25e990427c0c07f3a48b5e8af98a51d54688afab2a93a54832617f0c29ab4e8a","abstract_canon_sha256":"86afbada7c3cf0ff22e46ffd91b405ea4ba9cb155164ac3b0ab461a466058d04"},"schema_version":"1.0"},"canonical_sha256":"6a357c51fb5473bc005fac060f3510f6670afdf27fe52f895f30b84ed38b8943","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:57:59.385729Z","signature_b64":"ElA28ljZZifzbyWqz68W8Rw0m8u0jEpCJTATl2i36s6KWN+e69tcDLXUSHtZlb15CQMMhXCOQdhoh9vhv3BjBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a357c51fb5473bc005fac060f3510f6670afdf27fe52f895f30b84ed38b8943","last_reissued_at":"2026-07-05T04:57:59.385227Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:57:59.385227Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2209.07511","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":"2026-07-05T04:57:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lpCdDoN6IlBKKOHkGM5tZcs3s5O4PROJK9Sm3gzbQZF0lc418QTOgKEe+G89T9pxWwKPZXHIWOBqWkE/6CrfDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T08:21:51.939875Z"},"content_sha256":"0478780e95c59bf4acaa9f5083fdefebf57a1d1b68a9e0f856db25830a1eefcc","schema_version":"1.0","event_id":"sha256:0478780e95c59bf4acaa9f5083fdefebf57a1d1b68a9e0f856db25830a1eefcc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:NI2XYUP3KRZ3YAC7VQDA6NIQ6Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anima Anandkumar, Chaowei Xiao, De-An Huang, Manli Shu, Tom Goldstein, Weili Nie, Zhiding Yu","submitted_at":"2022-09-15T17:55:11Z","abstract_excerpt":"Pre-trained vision-language models (e.g., CLIP) have shown promising zero-shot generalization in many downstream tasks with properly designed text prompts. Instead of relying on hand-engineered prompts, recent works learn prompts using the training data from downstream tasks. While effective, training on domain-specific data reduces a model's generalization capability to unseen new domains. In this work, we propose test-time prompt tuning (TPT), a method that can learn adaptive prompts on the fly with a single test sample. For image classification, TPT optimizes the prompt by minimizing the en"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.07511","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/2209.07511/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":"2026-07-05T04:57:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0B6qNxlrcgj+/QkrCec/D01vEhL3tPDptc9oizb4BwQe5VQDy9N112vLbrNDJfFXbxXn7aorCzZLYKjJLUhgBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T08:21:51.940476Z"},"content_sha256":"5aa263ad6e45f1baa48af726caa4869dcd7af345d6a0c973d35029fef995660c","schema_version":"1.0","event_id":"sha256:5aa263ad6e45f1baa48af726caa4869dcd7af345d6a0c973d35029fef995660c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z/bundle.json","state_url":"https://pith.science/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z/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":"2026-08-09T08:21:51Z","links":{"resolver":"https://pith.science/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z","bundle":"https://pith.science/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z/bundle.json","state":"https://pith.science/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NI2XYUP3KRZ3YAC7VQDA6NIQ6Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:NI2XYUP3KRZ3YAC7VQDA6NIQ6Z","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":"86afbada7c3cf0ff22e46ffd91b405ea4ba9cb155164ac3b0ab461a466058d04","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-09-15T17:55:11Z","title_canon_sha256":"25e990427c0c07f3a48b5e8af98a51d54688afab2a93a54832617f0c29ab4e8a"},"schema_version":"1.0","source":{"id":"2209.07511","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2209.07511","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"arxiv_version","alias_value":"2209.07511v1","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.07511","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"pith_short_12","alias_value":"NI2XYUP3KRZ3","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"pith_short_16","alias_value":"NI2XYUP3KRZ3YAC7","created_at":"2026-07-05T04:57:59Z"},{"alias_kind":"pith_short_8","alias_value":"NI2XYUP3","created_at":"2026-07-05T04:57:59Z"}],"graph_snapshots":[{"event_id":"sha256:5aa263ad6e45f1baa48af726caa4869dcd7af345d6a0c973d35029fef995660c","target":"graph","created_at":"2026-07-05T04:57:59Z","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/2209.07511/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-trained vision-language models (e.g., CLIP) have shown promising zero-shot generalization in many downstream tasks with properly designed text prompts. Instead of relying on hand-engineered prompts, recent works learn prompts using the training data from downstream tasks. While effective, training on domain-specific data reduces a model's generalization capability to unseen new domains. In this work, we propose test-time prompt tuning (TPT), a method that can learn adaptive prompts on the fly with a single test sample. For image classification, TPT optimizes the prompt by minimizing the en","authors_text":"Anima Anandkumar, Chaowei Xiao, De-An Huang, Manli Shu, Tom Goldstein, Weili Nie, Zhiding Yu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-09-15T17:55:11Z","title":"Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.07511","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":"sha256:0478780e95c59bf4acaa9f5083fdefebf57a1d1b68a9e0f856db25830a1eefcc","target":"record","created_at":"2026-07-05T04:57:59Z","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":"86afbada7c3cf0ff22e46ffd91b405ea4ba9cb155164ac3b0ab461a466058d04","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-09-15T17:55:11Z","title_canon_sha256":"25e990427c0c07f3a48b5e8af98a51d54688afab2a93a54832617f0c29ab4e8a"},"schema_version":"1.0","source":{"id":"2209.07511","kind":"arxiv","version":1}},"canonical_sha256":"6a357c51fb5473bc005fac060f3510f6670afdf27fe52f895f30b84ed38b8943","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6a357c51fb5473bc005fac060f3510f6670afdf27fe52f895f30b84ed38b8943","first_computed_at":"2026-07-05T04:57:59.385227Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:57:59.385227Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ElA28ljZZifzbyWqz68W8Rw0m8u0jEpCJTATl2i36s6KWN+e69tcDLXUSHtZlb15CQMMhXCOQdhoh9vhv3BjBA==","signature_status":"signed_v1","signed_at":"2026-07-05T04:57:59.385729Z","signed_message":"canonical_sha256_bytes"},"source_id":"2209.07511","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0478780e95c59bf4acaa9f5083fdefebf57a1d1b68a9e0f856db25830a1eefcc","sha256:5aa263ad6e45f1baa48af726caa4869dcd7af345d6a0c973d35029fef995660c"],"state_sha256":"13b7ec6fc38c84694facffe83f8abd43f3739fa7641b157e7543dfd9b825392b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PmoeAN6Thu4DB0j5byziIICR6TxG6yPzJ9TcPqHgjkE+s6FTVf/OVQKbITPKT6Fe6/D1DyDUv6h2VxtrNmNlCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T08:21:51.944224Z","bundle_sha256":"7ef15ab95830c8e37185fc3a509e210e8d122ad55467ae28f6a953abd1ff5775"}}