{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:66SB556IYRXMWJBWO56YIQ3VCQ","short_pith_number":"pith:66SB556I","schema_version":"1.0","canonical_sha256":"f7a41ef7c8c46ecb2436777d844375142fcb7aced4e25c339038b29b34cb3cc7","source":{"kind":"arxiv","id":"2601.08828","version":2},"attestation_state":"computed","paper":{"title":"Motion Attribution for Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MM","cs.RO"],"primary_cat":"cs.CV","authors_text":"Antonio Torralba, Despoina Paschalidou, Jonathan Lorraine, Jun Gao, Laura Leal-Taix\\'e, Olga Russakovsky, Sanja Fidler, Xindi Wu","submitted_at":"2026-01-13T18:59:09Z","abstract_excerpt":"Despite the rapid progress of video generation models, the role of data in influencing motion is poorly understood. We present Motive (MOTIon attribution for Video gEneration), a motion-centric, gradient-based data attribution framework that scales to modern, large, high-quality video datasets and models. We use this to study which fine-tuning clips improve or degrade temporal dynamics. Motive isolates temporal dynamics from static appearance via motion-weighted loss masks, yielding efficient and scalable motion-specific influence computation. On text-to-video models, Motive identifies clips t"},"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":"2601.08828","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-01-13T18:59:09Z","cross_cats_sorted":["cs.AI","cs.LG","cs.MM","cs.RO"],"title_canon_sha256":"aad22eb50a76073b8e456c2303afe11332cb6450eb2a1a8e9e96a7fefdcab7f4","abstract_canon_sha256":"0e3e1e101ef74a9adeeada84adcfbb19c53ae8713c3cdb0c0cf1d1e6c912f860"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T01:16:03.749608Z","signature_b64":"O9LJ/B4jc7Lh93flvg49phZyrkTjk3kaikSMWVAXnJecelfNRmcP/2yYUojX5z0vNkKbPnpGU8cexFC0i7TwDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7a41ef7c8c46ecb2436777d844375142fcb7aced4e25c339038b29b34cb3cc7","last_reissued_at":"2026-07-07T01:16:03.748673Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T01:16:03.748673Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Motion Attribution for Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MM","cs.RO"],"primary_cat":"cs.CV","authors_text":"Antonio Torralba, Despoina Paschalidou, Jonathan Lorraine, Jun Gao, Laura Leal-Taix\\'e, Olga Russakovsky, Sanja Fidler, Xindi Wu","submitted_at":"2026-01-13T18:59:09Z","abstract_excerpt":"Despite the rapid progress of video generation models, the role of data in influencing motion is poorly understood. We present Motive (MOTIon attribution for Video gEneration), a motion-centric, gradient-based data attribution framework that scales to modern, large, high-quality video datasets and models. We use this to study which fine-tuning clips improve or degrade temporal dynamics. Motive isolates temporal dynamics from static appearance via motion-weighted loss masks, yielding efficient and scalable motion-specific influence computation. On text-to-video models, Motive identifies clips t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.08828","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/2601.08828/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":"2601.08828","created_at":"2026-07-07T01:16:03.748792+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.08828v2","created_at":"2026-07-07T01:16:03.748792+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.08828","created_at":"2026-07-07T01:16:03.748792+00:00"},{"alias_kind":"pith_short_12","alias_value":"66SB556IYRXM","created_at":"2026-07-07T01:16:03.748792+00:00"},{"alias_kind":"pith_short_16","alias_value":"66SB556IYRXMWJBW","created_at":"2026-07-07T01:16:03.748792+00:00"},{"alias_kind":"pith_short_8","alias_value":"66SB556I","created_at":"2026-07-07T01:16:03.748792+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.21489","citing_title":"Variance Reduction for Expectations with Diffusion Teachers","ref_index":89,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23878","citing_title":"LaMo: Self-Supervised Latent Motion Priors for Physical Realism in Video Generation","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21489","citing_title":"Variance Reduction for Expectations with Diffusion Teachers","ref_index":89,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ","json":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ.json","graph_json":"https://pith.science/api/pith-number/66SB556IYRXMWJBWO56YIQ3VCQ/graph.json","events_json":"https://pith.science/api/pith-number/66SB556IYRXMWJBWO56YIQ3VCQ/events.json","paper":"https://pith.science/paper/66SB556I"},"agent_actions":{"view_html":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ","download_json":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ.json","view_paper":"https://pith.science/paper/66SB556I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.08828&json=true","fetch_graph":"https://pith.science/api/pith-number/66SB556IYRXMWJBWO56YIQ3VCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/66SB556IYRXMWJBWO56YIQ3VCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ/action/storage_attestation","attest_author":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ/action/author_attestation","sign_citation":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ/action/citation_signature","submit_replication":"https://pith.science/pith/66SB556IYRXMWJBWO56YIQ3VCQ/action/replication_record"}},"created_at":"2026-07-07T01:16:03.748792+00:00","updated_at":"2026-07-07T01:16:03.748792+00:00"}