{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HZ5TM5YFCLSFURIJ5KPGLTCAFZ","short_pith_number":"pith:HZ5TM5YF","schema_version":"1.0","canonical_sha256":"3e7b36770512e45a4509ea9e65cc402e5c81eec3217dbbecc560298f49c5b699","source":{"kind":"arxiv","id":"2503.09279","version":1},"attestation_state":"computed","paper":{"title":"Cockatiel: Ensembling Synthetic and Human Preferenced Training for Detailed Video Caption","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Li, Luozheng Qin, Mengping Yang, Xiaomeng Yang, Zhiyu Tan","submitted_at":"2025-03-12T11:25:04Z","abstract_excerpt":"Video Detailed Captioning (VDC) is a crucial task for vision-language bridging, enabling fine-grained descriptions of complex video content. In this paper, we first comprehensively benchmark current state-of-the-art approaches and systematically identified two critical limitations: biased capability towards specific captioning aspect and misalignment with human preferences. To address these deficiencies, we propose Cockatiel, a novel three-stage training pipeline that ensembles synthetic and human-aligned training for improving VDC performance. In the first stage, we derive a scorer from a met"},"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":"2503.09279","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-12T11:25:04Z","cross_cats_sorted":[],"title_canon_sha256":"fb9efd6c0e03d1a04b5d6458f303d39afd2e36bd6d29ef930095eb7f7170a8e4","abstract_canon_sha256":"f3494c5679b4083a20f92a70827b1bb07d2f3851692e8999cf069382d0727906"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:29:53.209227Z","signature_b64":"GMSg1BIyLSi+kwusXNGNU5XiaUrMquKEfpBkESbN9Gwm2vSU0Dq1q/bDCQcmN2KPPpPW/cLxYa254+h3XdfICQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e7b36770512e45a4509ea9e65cc402e5c81eec3217dbbecc560298f49c5b699","last_reissued_at":"2026-07-05T10:29:53.208780Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:29:53.208780Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cockatiel: Ensembling Synthetic and Human Preferenced Training for Detailed Video Caption","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Li, Luozheng Qin, Mengping Yang, Xiaomeng Yang, Zhiyu Tan","submitted_at":"2025-03-12T11:25:04Z","abstract_excerpt":"Video Detailed Captioning (VDC) is a crucial task for vision-language bridging, enabling fine-grained descriptions of complex video content. In this paper, we first comprehensively benchmark current state-of-the-art approaches and systematically identified two critical limitations: biased capability towards specific captioning aspect and misalignment with human preferences. To address these deficiencies, we propose Cockatiel, a novel three-stage training pipeline that ensembles synthetic and human-aligned training for improving VDC performance. In the first stage, we derive a scorer from a met"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.09279","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/2503.09279/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":"2503.09279","created_at":"2026-07-05T10:29:53.208838+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.09279v1","created_at":"2026-07-05T10:29:53.208838+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.09279","created_at":"2026-07-05T10:29:53.208838+00:00"},{"alias_kind":"pith_short_12","alias_value":"HZ5TM5YFCLSF","created_at":"2026-07-05T10:29:53.208838+00:00"},{"alias_kind":"pith_short_16","alias_value":"HZ5TM5YFCLSFURIJ","created_at":"2026-07-05T10:29:53.208838+00:00"},{"alias_kind":"pith_short_8","alias_value":"HZ5TM5YF","created_at":"2026-07-05T10:29:53.208838+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/HZ5TM5YFCLSFURIJ5KPGLTCAFZ","json":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ.json","graph_json":"https://pith.science/api/pith-number/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/graph.json","events_json":"https://pith.science/api/pith-number/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/events.json","paper":"https://pith.science/paper/HZ5TM5YF"},"agent_actions":{"view_html":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ","download_json":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ.json","view_paper":"https://pith.science/paper/HZ5TM5YF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.09279&json=true","fetch_graph":"https://pith.science/api/pith-number/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/graph.json","fetch_events":"https://pith.science/api/pith-number/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/action/storage_attestation","attest_author":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/action/author_attestation","sign_citation":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/action/citation_signature","submit_replication":"https://pith.science/pith/HZ5TM5YFCLSFURIJ5KPGLTCAFZ/action/replication_record"}},"created_at":"2026-07-05T10:29:53.208838+00:00","updated_at":"2026-07-05T10:29:53.208838+00:00"}