{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:3EIDTBYIGG5U5KPSOYGG33JI62","short_pith_number":"pith:3EIDTBYI","canonical_record":{"source":{"id":"2507.15480","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-21T10:35:32Z","cross_cats_sorted":[],"title_canon_sha256":"6b3b9946e7cc1c5c5882e79bc6f9c788571bb748806e1ef9ea560880a00586ba","abstract_canon_sha256":"a88f77ced9ce11421392834a9d498fd125c5295acbd0b9830bf02769c37bfa90"},"schema_version":"1.0"},"canonical_sha256":"d91039870831bb4ea9f2760c6ded28f6a323a53c9010dc5abcbb82b30924fe36","source":{"kind":"arxiv","id":"2507.15480","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.15480","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"arxiv_version","alias_value":"2507.15480v2","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15480","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"pith_short_12","alias_value":"3EIDTBYIGG5U","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"pith_short_16","alias_value":"3EIDTBYIGG5U5KPS","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"pith_short_8","alias_value":"3EIDTBYI","created_at":"2026-07-05T11:44:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:3EIDTBYIGG5U5KPSOYGG33JI62","target":"record","payload":{"canonical_record":{"source":{"id":"2507.15480","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-21T10:35:32Z","cross_cats_sorted":[],"title_canon_sha256":"6b3b9946e7cc1c5c5882e79bc6f9c788571bb748806e1ef9ea560880a00586ba","abstract_canon_sha256":"a88f77ced9ce11421392834a9d498fd125c5295acbd0b9830bf02769c37bfa90"},"schema_version":"1.0"},"canonical_sha256":"d91039870831bb4ea9f2760c6ded28f6a323a53c9010dc5abcbb82b30924fe36","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:09.474191Z","signature_b64":"nQBjgXzKaCLGkMA+r+2M2FmOgmYKlhmecaMGHgzTWwLmNUg3v9ZjZmV191m7Id7c41R5kO+7qDJvC4jM4v5VBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d91039870831bb4ea9f2760c6ded28f6a323a53c9010dc5abcbb82b30924fe36","last_reissued_at":"2026-07-05T11:44:09.473672Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:09.473672Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.15480","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:44:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FaDgTg6Fd5ABVhWom2y8Y5zzkTEeEeVlZ8TdIpymYDIB/uN4cNE8OW42Ql8Xs4+XybfR/wchDzP2BSn8cLVlCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:22:00.600859Z"},"content_sha256":"8abe629538da06bca90e7248494d6159d43c7e2569b5a7072eb5518ed2ab3417","schema_version":"1.0","event_id":"sha256:8abe629538da06bca90e7248494d6159d43c7e2569b5a7072eb5518ed2ab3417"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:3EIDTBYIGG5U5KPSOYGG33JI62","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"One Last Attention for Your Vision-Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ghazi Shazan Ahmad, Liang Chen, Lingqiao Liu, Tianjun Yao, Zhiqiang Shen","submitted_at":"2025-07-21T10:35:32Z","abstract_excerpt":"Pretrained vision-language models (VLMs), such as CLIP, achieve remarkable zero-shot performance, yet their downstream potential hinges on effective fine-tuning. Most adaptation methods typically focus on refining representation from separate modalities (text or vision) but neglect the critical role of their fused representations in the decision-making process, \\emph{\\ie} rational matrix that drives the final prediction. To bridge the gap, we propose a simple yet effective \\textbf{R}ational \\textbf{Ada}ptaion ({RAda}) to explicitly exploit the final fused representation during fine-tuning. RAd"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15480","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/2507.15480/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":"2026-07-05T11:44:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V7EoMwMTwQ811Wj2nsaYHK9Ae3ahRdi6QZNp2P+/Lzca74MRfqd3Oorue//yLxQ4TcRiOUwg9NmxZJY1MJnuAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:22:00.601819Z"},"content_sha256":"ddcd7a271b2f10b3106b3e3bf4dffb96f8425b4a67f508e23a05d940fe941c5c","schema_version":"1.0","event_id":"sha256:ddcd7a271b2f10b3106b3e3bf4dffb96f8425b4a67f508e23a05d940fe941c5c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3EIDTBYIGG5U5KPSOYGG33JI62/bundle.json","state_url":"https://pith.science/pith/3EIDTBYIGG5U5KPSOYGG33JI62/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3EIDTBYIGG5U5KPSOYGG33JI62/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":"2026-08-04T16:22:00Z","links":{"resolver":"https://pith.science/pith/3EIDTBYIGG5U5KPSOYGG33JI62","bundle":"https://pith.science/pith/3EIDTBYIGG5U5KPSOYGG33JI62/bundle.json","state":"https://pith.science/pith/3EIDTBYIGG5U5KPSOYGG33JI62/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3EIDTBYIGG5U5KPSOYGG33JI62/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3EIDTBYIGG5U5KPSOYGG33JI62","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":"a88f77ced9ce11421392834a9d498fd125c5295acbd0b9830bf02769c37bfa90","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-21T10:35:32Z","title_canon_sha256":"6b3b9946e7cc1c5c5882e79bc6f9c788571bb748806e1ef9ea560880a00586ba"},"schema_version":"1.0","source":{"id":"2507.15480","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.15480","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"arxiv_version","alias_value":"2507.15480v2","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15480","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"pith_short_12","alias_value":"3EIDTBYIGG5U","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"pith_short_16","alias_value":"3EIDTBYIGG5U5KPS","created_at":"2026-07-05T11:44:09Z"},{"alias_kind":"pith_short_8","alias_value":"3EIDTBYI","created_at":"2026-07-05T11:44:09Z"}],"graph_snapshots":[{"event_id":"sha256:ddcd7a271b2f10b3106b3e3bf4dffb96f8425b4a67f508e23a05d940fe941c5c","target":"graph","created_at":"2026-07-05T11:44:09Z","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/2507.15480/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pretrained vision-language models (VLMs), such as CLIP, achieve remarkable zero-shot performance, yet their downstream potential hinges on effective fine-tuning. Most adaptation methods typically focus on refining representation from separate modalities (text or vision) but neglect the critical role of their fused representations in the decision-making process, \\emph{\\ie} rational matrix that drives the final prediction. To bridge the gap, we propose a simple yet effective \\textbf{R}ational \\textbf{Ada}ptaion ({RAda}) to explicitly exploit the final fused representation during fine-tuning. RAd","authors_text":"Ghazi Shazan Ahmad, Liang Chen, Lingqiao Liu, Tianjun Yao, Zhiqiang Shen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-21T10:35:32Z","title":"One Last Attention for Your Vision-Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15480","kind":"arxiv","version":2},"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":"sha256:8abe629538da06bca90e7248494d6159d43c7e2569b5a7072eb5518ed2ab3417","target":"record","created_at":"2026-07-05T11:44:09Z","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":"a88f77ced9ce11421392834a9d498fd125c5295acbd0b9830bf02769c37bfa90","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-21T10:35:32Z","title_canon_sha256":"6b3b9946e7cc1c5c5882e79bc6f9c788571bb748806e1ef9ea560880a00586ba"},"schema_version":"1.0","source":{"id":"2507.15480","kind":"arxiv","version":2}},"canonical_sha256":"d91039870831bb4ea9f2760c6ded28f6a323a53c9010dc5abcbb82b30924fe36","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d91039870831bb4ea9f2760c6ded28f6a323a53c9010dc5abcbb82b30924fe36","first_computed_at":"2026-07-05T11:44:09.473672Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:44:09.473672Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nQBjgXzKaCLGkMA+r+2M2FmOgmYKlhmecaMGHgzTWwLmNUg3v9ZjZmV191m7Id7c41R5kO+7qDJvC4jM4v5VBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:44:09.474191Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.15480","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8abe629538da06bca90e7248494d6159d43c7e2569b5a7072eb5518ed2ab3417","sha256:ddcd7a271b2f10b3106b3e3bf4dffb96f8425b4a67f508e23a05d940fe941c5c"],"state_sha256":"6bca7b21526c7ca31bde506fc81bc9c9b8a260b063f3caa8f5b74dcb5f1310eb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pTw85Av8jEMSXDVf413zhgij7QvGBWJWYZIRpovC7gJQLYBZeLfrgqGgUFZf4itL7jia85d2+uJ4Y+Ip+NQPDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T16:22:00.632388Z","bundle_sha256":"c46d6cc6c653cccc74fa400ca28ed7575e7919f445701504afe1b075980c1257"}}