{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:H6X6TFTAAPBDRUVJ5PDFJOOSAW","short_pith_number":"pith:H6X6TFTA","schema_version":"1.0","canonical_sha256":"3fafe9966003c238d2a9ebc654b9d205916ee8025f595a44dcae3952f9dba532","source":{"kind":"arxiv","id":"2402.13250","version":6},"attestation_state":"computed","paper":{"title":"Video ReCap: Recursive Captioning of Hour-Long Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gedas Bertasius, Lorenzo Torresani, Md Mohaiminul Islam, Ngan Ho, Tushar Nagarajan, Xitong Yang","submitted_at":"2024-02-20T18:58:54Z","abstract_excerpt":"Most video captioning models are designed to process short video clips of few seconds and output text describing low-level visual concepts (e.g., objects, scenes, atomic actions). However, most real-world videos last for minutes or hours and have a complex hierarchical structure spanning different temporal granularities. We propose Video ReCap, a recursive video captioning model that can process video inputs of dramatically different lengths (from 1 second to 2 hours) and output video captions at multiple hierarchy levels. The recursive video-language architecture exploits the synergy between "},"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":"2402.13250","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-02-20T18:58:54Z","cross_cats_sorted":[],"title_canon_sha256":"e02869ecb1f52996a295bc2a647898b1b8c5669fdf0f981f4d94447220c89075","abstract_canon_sha256":"2811450cc209310e20db7af4acaffdd6b713f0d88239ed18e644b313ca0adb6f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:19:58.105392Z","signature_b64":"8oUFYscVZkhv90uvhX/u6YZKJ6tRhQmkEQU6Xgv7Q6h4HL8EDLYVsk++u3f+nJrCDqI4k/6lF1754LVdHmkUDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3fafe9966003c238d2a9ebc654b9d205916ee8025f595a44dcae3952f9dba532","last_reissued_at":"2026-07-05T08:19:58.104912Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:19:58.104912Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Video ReCap: Recursive Captioning of Hour-Long Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gedas Bertasius, Lorenzo Torresani, Md Mohaiminul Islam, Ngan Ho, Tushar Nagarajan, Xitong Yang","submitted_at":"2024-02-20T18:58:54Z","abstract_excerpt":"Most video captioning models are designed to process short video clips of few seconds and output text describing low-level visual concepts (e.g., objects, scenes, atomic actions). However, most real-world videos last for minutes or hours and have a complex hierarchical structure spanning different temporal granularities. We propose Video ReCap, a recursive video captioning model that can process video inputs of dramatically different lengths (from 1 second to 2 hours) and output video captions at multiple hierarchy levels. The recursive video-language architecture exploits the synergy between "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.13250","kind":"arxiv","version":6},"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/2402.13250/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":"2402.13250","created_at":"2026-07-05T08:19:58.104970+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.13250v6","created_at":"2026-07-05T08:19:58.104970+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.13250","created_at":"2026-07-05T08:19:58.104970+00:00"},{"alias_kind":"pith_short_12","alias_value":"H6X6TFTAAPBD","created_at":"2026-07-05T08:19:58.104970+00:00"},{"alias_kind":"pith_short_16","alias_value":"H6X6TFTAAPBDRUVJ","created_at":"2026-07-05T08:19:58.104970+00:00"},{"alias_kind":"pith_short_8","alias_value":"H6X6TFTA","created_at":"2026-07-05T08:19:58.104970+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13141","citing_title":"Rethinking RAG in Long Videos: What to Retrieve and How to Use It?","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.16009","citing_title":"Bridging the Usability Gap: Lessons from Interpreting Studies for Machine Interpreting Design","ref_index":121,"is_internal_anchor":false},{"citing_arxiv_id":"2407.03320","citing_title":"InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2410.02713","citing_title":"LLaVA-Video: Video Instruction Tuning With Synthetic Data","ref_index":215,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW","json":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW.json","graph_json":"https://pith.science/api/pith-number/H6X6TFTAAPBDRUVJ5PDFJOOSAW/graph.json","events_json":"https://pith.science/api/pith-number/H6X6TFTAAPBDRUVJ5PDFJOOSAW/events.json","paper":"https://pith.science/paper/H6X6TFTA"},"agent_actions":{"view_html":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW","download_json":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW.json","view_paper":"https://pith.science/paper/H6X6TFTA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.13250&json=true","fetch_graph":"https://pith.science/api/pith-number/H6X6TFTAAPBDRUVJ5PDFJOOSAW/graph.json","fetch_events":"https://pith.science/api/pith-number/H6X6TFTAAPBDRUVJ5PDFJOOSAW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW/action/storage_attestation","attest_author":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW/action/author_attestation","sign_citation":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW/action/citation_signature","submit_replication":"https://pith.science/pith/H6X6TFTAAPBDRUVJ5PDFJOOSAW/action/replication_record"}},"created_at":"2026-07-05T08:19:58.104970+00:00","updated_at":"2026-07-05T08:19:58.104970+00:00"}