{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LN4D2RQJVWTQJK5JEK63GWTSBK","short_pith_number":"pith:LN4D2RQJ","schema_version":"1.0","canonical_sha256":"5b783d4609ada704aba922bdb35a720a8255042d7de0353476d1972f5b94303f","source":{"kind":"arxiv","id":"2607.02963","version":1},"attestation_state":"computed","paper":{"title":"Parallelized Autoregressive Decoding for Omni-Modal Dense Video Captioning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.MM"],"primary_cat":"cs.CV","authors_text":"Chen Gao, Hwee Tou Ng, Mike Zheng Shou, Siyi Jiao, Wenzheng Zeng","submitted_at":"2026-07-03T05:13:19Z","abstract_excerpt":"Dense video captioning aims to generate temporally grounded descriptions of video events, benefiting both event-level video understanding and generation. In this domain, autoregressive video large language models have emerged as a prevalent paradigm due to their strong generative and cross-modal modeling capacity. However, generating dense captions under the token-by-token paradigm severely limits inference efficiency and hinders scalability as video length and event density increase. In this work, we propose a parallelized autoregressive framework that not only improves generation efficiency "},"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":"2607.02963","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-03T05:13:19Z","cross_cats_sorted":["cs.AI","cs.MM"],"title_canon_sha256":"1d71f23e12b66f772ce2d99697245c03ec5d5306c7ac55c3a3b6a7a43c50f02f","abstract_canon_sha256":"876baeca44945b517fc76cf31000b60cdda023c0afac5295d14d1f2b44a76a10"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T01:16:38.010389Z","signature_b64":"WQnItds1dzX+Wu7Y+T4goAsZnY5ZlVYi4DceWwQfS/2pDxkJNbtcFboUaUJkFKsvmXLR5h00OAHSol2TdFlzAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b783d4609ada704aba922bdb35a720a8255042d7de0353476d1972f5b94303f","last_reissued_at":"2026-07-07T01:16:38.009954Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T01:16:38.009954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Parallelized Autoregressive Decoding for Omni-Modal Dense Video Captioning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.MM"],"primary_cat":"cs.CV","authors_text":"Chen Gao, Hwee Tou Ng, Mike Zheng Shou, Siyi Jiao, Wenzheng Zeng","submitted_at":"2026-07-03T05:13:19Z","abstract_excerpt":"Dense video captioning aims to generate temporally grounded descriptions of video events, benefiting both event-level video understanding and generation. In this domain, autoregressive video large language models have emerged as a prevalent paradigm due to their strong generative and cross-modal modeling capacity. However, generating dense captions under the token-by-token paradigm severely limits inference efficiency and hinders scalability as video length and event density increase. In this work, we propose a parallelized autoregressive framework that not only improves generation efficiency "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02963","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/2607.02963/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":"2607.02963","created_at":"2026-07-07T01:16:38.010022+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02963v1","created_at":"2026-07-07T01:16:38.010022+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02963","created_at":"2026-07-07T01:16:38.010022+00:00"},{"alias_kind":"pith_short_12","alias_value":"LN4D2RQJVWTQ","created_at":"2026-07-07T01:16:38.010022+00:00"},{"alias_kind":"pith_short_16","alias_value":"LN4D2RQJVWTQJK5J","created_at":"2026-07-07T01:16:38.010022+00:00"},{"alias_kind":"pith_short_8","alias_value":"LN4D2RQJ","created_at":"2026-07-07T01:16:38.010022+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK","json":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK.json","graph_json":"https://pith.science/api/pith-number/LN4D2RQJVWTQJK5JEK63GWTSBK/graph.json","events_json":"https://pith.science/api/pith-number/LN4D2RQJVWTQJK5JEK63GWTSBK/events.json","paper":"https://pith.science/paper/LN4D2RQJ"},"agent_actions":{"view_html":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK","download_json":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK.json","view_paper":"https://pith.science/paper/LN4D2RQJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02963&json=true","fetch_graph":"https://pith.science/api/pith-number/LN4D2RQJVWTQJK5JEK63GWTSBK/graph.json","fetch_events":"https://pith.science/api/pith-number/LN4D2RQJVWTQJK5JEK63GWTSBK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK/action/storage_attestation","attest_author":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK/action/author_attestation","sign_citation":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK/action/citation_signature","submit_replication":"https://pith.science/pith/LN4D2RQJVWTQJK5JEK63GWTSBK/action/replication_record"}},"created_at":"2026-07-07T01:16:38.010022+00:00","updated_at":"2026-07-07T01:16:38.010022+00:00"}