{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:22YEYZH4H6S3S6ROAY7SBUE7IL","short_pith_number":"pith:22YEYZH4","schema_version":"1.0","canonical_sha256":"d6b04c64fc3fa5b97a2e063f20d09f42e026fe6a333df2c6696582246d9b8698","source":{"kind":"arxiv","id":"2507.13609","version":1},"attestation_state":"computed","paper":{"title":"CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Hao Niu, Julio Vizcarra, Mori Kurokawa, Yanan Wang, Zhi Li","submitted_at":"2025-07-18T02:29:19Z","abstract_excerpt":"Despite recent progress in video large language models (VideoLLMs), a key open challenge remains: how to equip models with chain-of-thought (CoT) reasoning abilities grounded in fine-grained object-level video understanding. Existing instruction-tuned models, such as the Qwen and LLaVA series, are trained on high-level video-text pairs, often lacking structured annotations necessary for compositional, step-by-step reasoning. We propose CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks, a new framework that decomposes complex video questions of existing datasets (e.g., NeXT-QA, STA"},"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":"2507.13609","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-18T02:29:19Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"953ba2ec944f65cd599f71be1bb3ca566b2b8e6892c427d3dc0dae997c48906c","abstract_canon_sha256":"ed07bb8f8c867c45ad6d993b0c5cd44270e97844a2d7cafa04f850d077399e95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:39:20.131357Z","signature_b64":"G0XAQjBfdrHt0P0WQBQhca6dtAfFJCV2j+9Z9CVN5/rHBOa3s+tsId/bdLqW2IRGXEWdWZIJ4SdQE2S1/+xLDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6b04c64fc3fa5b97a2e063f20d09f42e026fe6a333df2c6696582246d9b8698","last_reissued_at":"2026-07-05T11:39:20.130851Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:39:20.130851Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Hao Niu, Julio Vizcarra, Mori Kurokawa, Yanan Wang, Zhi Li","submitted_at":"2025-07-18T02:29:19Z","abstract_excerpt":"Despite recent progress in video large language models (VideoLLMs), a key open challenge remains: how to equip models with chain-of-thought (CoT) reasoning abilities grounded in fine-grained object-level video understanding. Existing instruction-tuned models, such as the Qwen and LLaVA series, are trained on high-level video-text pairs, often lacking structured annotations necessary for compositional, step-by-step reasoning. We propose CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks, a new framework that decomposes complex video questions of existing datasets (e.g., NeXT-QA, STA"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.13609","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/2507.13609/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":"2507.13609","created_at":"2026-07-05T11:39:20.130916+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.13609v1","created_at":"2026-07-05T11:39:20.130916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.13609","created_at":"2026-07-05T11:39:20.130916+00:00"},{"alias_kind":"pith_short_12","alias_value":"22YEYZH4H6S3","created_at":"2026-07-05T11:39:20.130916+00:00"},{"alias_kind":"pith_short_16","alias_value":"22YEYZH4H6S3S6RO","created_at":"2026-07-05T11:39:20.130916+00:00"},{"alias_kind":"pith_short_8","alias_value":"22YEYZH4","created_at":"2026-07-05T11:39:20.130916+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.08209","citing_title":"OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL","json":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL.json","graph_json":"https://pith.science/api/pith-number/22YEYZH4H6S3S6ROAY7SBUE7IL/graph.json","events_json":"https://pith.science/api/pith-number/22YEYZH4H6S3S6ROAY7SBUE7IL/events.json","paper":"https://pith.science/paper/22YEYZH4"},"agent_actions":{"view_html":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL","download_json":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL.json","view_paper":"https://pith.science/paper/22YEYZH4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.13609&json=true","fetch_graph":"https://pith.science/api/pith-number/22YEYZH4H6S3S6ROAY7SBUE7IL/graph.json","fetch_events":"https://pith.science/api/pith-number/22YEYZH4H6S3S6ROAY7SBUE7IL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL/action/storage_attestation","attest_author":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL/action/author_attestation","sign_citation":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL/action/citation_signature","submit_replication":"https://pith.science/pith/22YEYZH4H6S3S6ROAY7SBUE7IL/action/replication_record"}},"created_at":"2026-07-05T11:39:20.130916+00:00","updated_at":"2026-07-05T11:39:20.130916+00:00"}