{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:G2DTWLHTDYEHDYQY4TIODTDGWJ","short_pith_number":"pith:G2DTWLHT","schema_version":"1.0","canonical_sha256":"36873b2cf31e0871e218e4d0e1cc66b24400be492c3ebc4b07da4ceed020e409","source":{"kind":"arxiv","id":"2412.01694","version":2},"attestation_state":"computed","paper":{"title":"Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Qirui Chen, Shangzhe Di, Weidi Xie, Yudi Shi","submitted_at":"2024-12-02T16:37:50Z","abstract_excerpt":"This paper tackles the problem of video question answering (VideoQA), a task that often requires multi-step reasoning and a profound understanding of spatial-temporal dynamics. While large video-language models perform well on benchmarks, they often lack explainability and spatial-temporal grounding. In this paper, we propose Agent-of-Thoughts Distillation (AoTD), a method that enhances models by incorporating automatically generated Chain-of-Thoughts (CoTs) into the instruction-tuning process. Specifically, we leverage an agent-based system to decompose complex questions into sub-tasks, and a"},"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":"2412.01694","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T16:37:50Z","cross_cats_sorted":[],"title_canon_sha256":"34fd6c28eb4c2616b2b66ad40d4bb85cab37d35ed2b82b21fb83a55be6bb630e","abstract_canon_sha256":"6e8281f41e90cec38bfb74673809b0d21d842a2496ac1717cadea49f5f37db81"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:45.090271Z","signature_b64":"oazG/mG9VHDNEIU1aIqzNtDIPgwelqAjiLMFomMLBwWTljxQJE9W0NycCp4vEbr0Qtg0N3Rf3yLd/cw1t35oAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"36873b2cf31e0871e218e4d0e1cc66b24400be492c3ebc4b07da4ceed020e409","last_reissued_at":"2026-07-05T10:13:45.089693Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:45.089693Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Qirui Chen, Shangzhe Di, Weidi Xie, Yudi Shi","submitted_at":"2024-12-02T16:37:50Z","abstract_excerpt":"This paper tackles the problem of video question answering (VideoQA), a task that often requires multi-step reasoning and a profound understanding of spatial-temporal dynamics. While large video-language models perform well on benchmarks, they often lack explainability and spatial-temporal grounding. In this paper, we propose Agent-of-Thoughts Distillation (AoTD), a method that enhances models by incorporating automatically generated Chain-of-Thoughts (CoTs) into the instruction-tuning process. Specifically, we leverage an agent-based system to decompose complex questions into sub-tasks, and a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01694","kind":"arxiv","version":2},"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/2412.01694/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":"2412.01694","created_at":"2026-07-05T10:13:45.089780+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.01694v2","created_at":"2026-07-05T10:13:45.089780+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01694","created_at":"2026-07-05T10:13:45.089780+00:00"},{"alias_kind":"pith_short_12","alias_value":"G2DTWLHTDYEH","created_at":"2026-07-05T10:13:45.089780+00:00"},{"alias_kind":"pith_short_16","alias_value":"G2DTWLHTDYEHDYQY","created_at":"2026-07-05T10:13:45.089780+00:00"},{"alias_kind":"pith_short_8","alias_value":"G2DTWLHT","created_at":"2026-07-05T10:13:45.089780+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23216","citing_title":"CaST-Bench: Benchmarking Causal Chain-Grounded Spatio-Temporal Reasoning for Video Question Answering","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00867","citing_title":"EFlow: Learning Evidence Flow for Long-Video Reasoning with Adaptive Reflection","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23216","citing_title":"CaST-Bench: Benchmarking Causal Chain-Grounded Spatio-Temporal Reasoning for Video Question Answering","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ","json":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ.json","graph_json":"https://pith.science/api/pith-number/G2DTWLHTDYEHDYQY4TIODTDGWJ/graph.json","events_json":"https://pith.science/api/pith-number/G2DTWLHTDYEHDYQY4TIODTDGWJ/events.json","paper":"https://pith.science/paper/G2DTWLHT"},"agent_actions":{"view_html":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ","download_json":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ.json","view_paper":"https://pith.science/paper/G2DTWLHT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.01694&json=true","fetch_graph":"https://pith.science/api/pith-number/G2DTWLHTDYEHDYQY4TIODTDGWJ/graph.json","fetch_events":"https://pith.science/api/pith-number/G2DTWLHTDYEHDYQY4TIODTDGWJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ/action/storage_attestation","attest_author":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ/action/author_attestation","sign_citation":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ/action/citation_signature","submit_replication":"https://pith.science/pith/G2DTWLHTDYEHDYQY4TIODTDGWJ/action/replication_record"}},"created_at":"2026-07-05T10:13:45.089780+00:00","updated_at":"2026-07-05T10:13:45.089780+00:00"}