{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6SZV2WS74EVQ6F77UOTEEGHOGQ","short_pith_number":"pith:6SZV2WS7","schema_version":"1.0","canonical_sha256":"f4b35d5a5fe12b0f17ffa3a64218ee340d00eb4bcfce765387fbc73224aae4a4","source":{"kind":"arxiv","id":"2305.03204","version":1},"attestation_state":"computed","paper":{"title":"VideoOFA: Two-Stage Pre-Training for Video-to-Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Barlas O\\u{g}uz, Lili Yu, Wenhan Xiong, Wen-tau Yih, Xilun Chen, Yashar Mehdad","submitted_at":"2023-05-04T23:27:21Z","abstract_excerpt":"We propose a new two-stage pre-training framework for video-to-text generation tasks such as video captioning and video question answering: A generative encoder-decoder model is first jointly pre-trained on massive image-text data to learn fundamental vision-language concepts, and then adapted to video data in an intermediate video-text pre-training stage to learn video-specific skills such as spatio-temporal reasoning. As a result, our VideoOFA model achieves new state-of-the-art performance on four Video Captioning benchmarks, beating prior art by an average of 9.7 points in CIDEr score. It "},"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":"2305.03204","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-04T23:27:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"50a4d8d5d2a5e18bbd045bcaff67971d58fc8d4526064291c17933e9603d38b0","abstract_canon_sha256":"f6b00e73cf1c3e2c9c47bbe7f116d1c73bbb4738e776cd0e6cab3736ef1e1266"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:07:11.923142Z","signature_b64":"swzdSs0J/J91YCwGQX3uthCBGDiQiMjHZyahRDIWL5izTT6sKrQ/a124r1kX6I+vT0nf5NqYomFNUJKlN7edAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4b35d5a5fe12b0f17ffa3a64218ee340d00eb4bcfce765387fbc73224aae4a4","last_reissued_at":"2026-07-05T06:07:11.922669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:07:11.922669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VideoOFA: Two-Stage Pre-Training for Video-to-Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Barlas O\\u{g}uz, Lili Yu, Wenhan Xiong, Wen-tau Yih, Xilun Chen, Yashar Mehdad","submitted_at":"2023-05-04T23:27:21Z","abstract_excerpt":"We propose a new two-stage pre-training framework for video-to-text generation tasks such as video captioning and video question answering: A generative encoder-decoder model is first jointly pre-trained on massive image-text data to learn fundamental vision-language concepts, and then adapted to video data in an intermediate video-text pre-training stage to learn video-specific skills such as spatio-temporal reasoning. As a result, our VideoOFA model achieves new state-of-the-art performance on four Video Captioning benchmarks, beating prior art by an average of 9.7 points in CIDEr score. It "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.03204","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/2305.03204/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":"2305.03204","created_at":"2026-07-05T06:07:11.922726+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.03204v1","created_at":"2026-07-05T06:07:11.922726+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.03204","created_at":"2026-07-05T06:07:11.922726+00:00"},{"alias_kind":"pith_short_12","alias_value":"6SZV2WS74EVQ","created_at":"2026-07-05T06:07:11.922726+00:00"},{"alias_kind":"pith_short_16","alias_value":"6SZV2WS74EVQ6F77","created_at":"2026-07-05T06:07:11.922726+00:00"},{"alias_kind":"pith_short_8","alias_value":"6SZV2WS7","created_at":"2026-07-05T06:07:11.922726+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.15220","citing_title":"SyncFlow: Toward Temporally Aligned Joint Audio-Video Generation from Text","ref_index":2020,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ","json":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ.json","graph_json":"https://pith.science/api/pith-number/6SZV2WS74EVQ6F77UOTEEGHOGQ/graph.json","events_json":"https://pith.science/api/pith-number/6SZV2WS74EVQ6F77UOTEEGHOGQ/events.json","paper":"https://pith.science/paper/6SZV2WS7"},"agent_actions":{"view_html":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ","download_json":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ.json","view_paper":"https://pith.science/paper/6SZV2WS7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.03204&json=true","fetch_graph":"https://pith.science/api/pith-number/6SZV2WS74EVQ6F77UOTEEGHOGQ/graph.json","fetch_events":"https://pith.science/api/pith-number/6SZV2WS74EVQ6F77UOTEEGHOGQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ/action/storage_attestation","attest_author":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ/action/author_attestation","sign_citation":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ/action/citation_signature","submit_replication":"https://pith.science/pith/6SZV2WS74EVQ6F77UOTEEGHOGQ/action/replication_record"}},"created_at":"2026-07-05T06:07:11.922726+00:00","updated_at":"2026-07-05T06:07:11.922726+00:00"}