{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:UJYIT7GOBLPEO6VOSYR5OGSODO","short_pith_number":"pith:UJYIT7GO","canonical_record":{"source":{"id":"2211.00575","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-11-01T16:36:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"36ebf2727b95b2f9492e90bd0f6ab36fec313db0abffd584d5075ef7753fb70c","abstract_canon_sha256":"c60965cd8979e43e02fdcd56618a229283c5cbb90e70edd623746daabbc07a4c"},"schema_version":"1.0"},"canonical_sha256":"a27089fcce0ade477aae9623d71a4e1bb58a6175385e01d0d4734d098fc957f0","source":{"kind":"arxiv","id":"2211.00575","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.00575","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"arxiv_version","alias_value":"2211.00575v1","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.00575","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"pith_short_12","alias_value":"UJYIT7GOBLPE","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"pith_short_16","alias_value":"UJYIT7GOBLPEO6VO","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"pith_short_8","alias_value":"UJYIT7GO","created_at":"2026-07-05T06:59:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:UJYIT7GOBLPEO6VOSYR5OGSODO","target":"record","payload":{"canonical_record":{"source":{"id":"2211.00575","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-11-01T16:36:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"36ebf2727b95b2f9492e90bd0f6ab36fec313db0abffd584d5075ef7753fb70c","abstract_canon_sha256":"c60965cd8979e43e02fdcd56618a229283c5cbb90e70edd623746daabbc07a4c"},"schema_version":"1.0"},"canonical_sha256":"a27089fcce0ade477aae9623d71a4e1bb58a6175385e01d0d4734d098fc957f0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:59:50.305351Z","signature_b64":"A3r3+ShmI83rL5aWTDx+LkPd8zOOh7PTq7ccxYj89WoxdQ02cd9UDew0ehfMMlmoXRN22kzWnMkA+vzVT5AkCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a27089fcce0ade477aae9623d71a4e1bb58a6175385e01d0d4734d098fc957f0","last_reissued_at":"2026-07-05T06:59:50.304859Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:59:50.304859Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2211.00575","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":"2026-07-05T06:59:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gt2ky4CcLIRL6iTVnakg6UVfA7dTMlf3PHepVnOPGcHq3VNU2uNOgy8CRvnanXk3oBMaaivUhuG4TkigHNWwCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T04:45:15.412169Z"},"content_sha256":"394cbd20c787ca0abfa681506f5abb416c23f3aaf627c6e69e7fb39cd715af39","schema_version":"1.0","event_id":"sha256:394cbd20c787ca0abfa681506f5abb416c23f3aaf627c6e69e7fb39cd715af39"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:UJYIT7GOBLPEO6VOSYR5OGSODO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Text-Only Training for Image Captioning using Noise-Injected CLIP","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Amir Globerson, David Nukrai, Ron Mokady","submitted_at":"2022-11-01T16:36:01Z","abstract_excerpt":"We consider the task of image-captioning using only the CLIP model and additional text data at training time, and no additional captioned images. Our approach relies on the fact that CLIP is trained to make visual and textual embeddings similar. Therefore, we only need to learn how to translate CLIP textual embeddings back into text, and we can learn how to do this by learning a decoder for the frozen CLIP text encoder using only text. We argue that this intuition is \"almost correct\" because of a gap between the embedding spaces, and propose to rectify this via noise injection during training."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.00575","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/2211.00575/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":"2026-07-05T06:59:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+skggJRS9XKaDRDI4j/AjokS0N+A9GtXegaBOCW/9+e4rfw0ddD62e/b0vXD1Wv9JWcFWvuZ3NXZFu0O/jv4AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T04:45:15.412726Z"},"content_sha256":"94f75610bea8a89e2cba822a84ac490265e76dcb3c7a39c2cee7a7958b3e8dfd","schema_version":"1.0","event_id":"sha256:94f75610bea8a89e2cba822a84ac490265e76dcb3c7a39c2cee7a7958b3e8dfd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UJYIT7GOBLPEO6VOSYR5OGSODO/bundle.json","state_url":"https://pith.science/pith/UJYIT7GOBLPEO6VOSYR5OGSODO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UJYIT7GOBLPEO6VOSYR5OGSODO/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":"2026-08-08T04:45:15Z","links":{"resolver":"https://pith.science/pith/UJYIT7GOBLPEO6VOSYR5OGSODO","bundle":"https://pith.science/pith/UJYIT7GOBLPEO6VOSYR5OGSODO/bundle.json","state":"https://pith.science/pith/UJYIT7GOBLPEO6VOSYR5OGSODO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UJYIT7GOBLPEO6VOSYR5OGSODO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:UJYIT7GOBLPEO6VOSYR5OGSODO","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":"c60965cd8979e43e02fdcd56618a229283c5cbb90e70edd623746daabbc07a4c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-11-01T16:36:01Z","title_canon_sha256":"36ebf2727b95b2f9492e90bd0f6ab36fec313db0abffd584d5075ef7753fb70c"},"schema_version":"1.0","source":{"id":"2211.00575","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.00575","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"arxiv_version","alias_value":"2211.00575v1","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.00575","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"pith_short_12","alias_value":"UJYIT7GOBLPE","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"pith_short_16","alias_value":"UJYIT7GOBLPEO6VO","created_at":"2026-07-05T06:59:50Z"},{"alias_kind":"pith_short_8","alias_value":"UJYIT7GO","created_at":"2026-07-05T06:59:50Z"}],"graph_snapshots":[{"event_id":"sha256:94f75610bea8a89e2cba822a84ac490265e76dcb3c7a39c2cee7a7958b3e8dfd","target":"graph","created_at":"2026-07-05T06:59:50Z","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/2211.00575/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We consider the task of image-captioning using only the CLIP model and additional text data at training time, and no additional captioned images. Our approach relies on the fact that CLIP is trained to make visual and textual embeddings similar. Therefore, we only need to learn how to translate CLIP textual embeddings back into text, and we can learn how to do this by learning a decoder for the frozen CLIP text encoder using only text. We argue that this intuition is \"almost correct\" because of a gap between the embedding spaces, and propose to rectify this via noise injection during training.","authors_text":"Amir Globerson, David Nukrai, Ron Mokady","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-11-01T16:36:01Z","title":"Text-Only Training for Image Captioning using Noise-Injected CLIP"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.00575","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":"sha256:394cbd20c787ca0abfa681506f5abb416c23f3aaf627c6e69e7fb39cd715af39","target":"record","created_at":"2026-07-05T06:59:50Z","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":"c60965cd8979e43e02fdcd56618a229283c5cbb90e70edd623746daabbc07a4c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-11-01T16:36:01Z","title_canon_sha256":"36ebf2727b95b2f9492e90bd0f6ab36fec313db0abffd584d5075ef7753fb70c"},"schema_version":"1.0","source":{"id":"2211.00575","kind":"arxiv","version":1}},"canonical_sha256":"a27089fcce0ade477aae9623d71a4e1bb58a6175385e01d0d4734d098fc957f0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a27089fcce0ade477aae9623d71a4e1bb58a6175385e01d0d4734d098fc957f0","first_computed_at":"2026-07-05T06:59:50.304859Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:59:50.304859Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"A3r3+ShmI83rL5aWTDx+LkPd8zOOh7PTq7ccxYj89WoxdQ02cd9UDew0ehfMMlmoXRN22kzWnMkA+vzVT5AkCg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:59:50.305351Z","signed_message":"canonical_sha256_bytes"},"source_id":"2211.00575","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:394cbd20c787ca0abfa681506f5abb416c23f3aaf627c6e69e7fb39cd715af39","sha256:94f75610bea8a89e2cba822a84ac490265e76dcb3c7a39c2cee7a7958b3e8dfd"],"state_sha256":"ccd7bf254fa5a8071067058757a92349c2fb158e676a65e838776f80a64a7ff4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"X6b5FuVXJQ66jWod3oGhzIiaD0gZwoZaKtGCezrGwVGGN59aviuOQMvOjJtCwR/GPEwYuuKVQucObp8nkZgvDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T04:45:15.417817Z","bundle_sha256":"ef6534edf402b09fcbda40db344f94753b1758c994e0ec527a60885240e51683"}}