{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RGGC6JDB6VCPHQURQQZOQAO5F6","short_pith_number":"pith:RGGC6JDB","schema_version":"1.0","canonical_sha256":"898c2f2461f544f3c2918432e801dd2fb7f99c117056a967ab7c59f1719107d9","source":{"kind":"arxiv","id":"2412.07750","version":3},"attestation_state":"computed","paper":{"title":"Motion by Queries: Identity-Motion Trade-offs in Text-to-Video Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gal Chechik, Rinon Gal, Yoad Tewel, Yoni Kasten, Yuval Atzmon","submitted_at":"2024-12-10T18:49:39Z","abstract_excerpt":"Text-to-video diffusion models have shown remarkable progress in generating coherent video clips from textual descriptions. However, the interplay between motion, structure, and identity representations in these models remains under-explored. Here, we investigate how self-attention query (Q) features simultaneously govern motion, structure, and identity and examine the challenges arising when these representations interact. Our analysis reveals that Q affects not only layout, but that during denoising Q also has a strong effect on subject identity, making it hard to transfer motion without the"},"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":"2412.07750","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-10T18:49:39Z","cross_cats_sorted":[],"title_canon_sha256":"737911a0dbb94dab69096eaf7e69d279817927c3139a16c3b437249f08ad4daa","abstract_canon_sha256":"d7ce1c3c2ce029091ecf1efa45857b65b658aaf18fa4de9a4c9782c43ffb74b0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:17.748961Z","signature_b64":"RFbySOSSzd9s5UfXH/Whf7lUeKC2RimzsvIV5w+qMDAaU19qnPn7ZaKI0gavSIY2K6fGQkXuhYvNMBAvzaNVDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"898c2f2461f544f3c2918432e801dd2fb7f99c117056a967ab7c59f1719107d9","last_reissued_at":"2026-07-05T11:07:17.748487Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:17.748487Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Motion by Queries: Identity-Motion Trade-offs in Text-to-Video Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gal Chechik, Rinon Gal, Yoad Tewel, Yoni Kasten, Yuval Atzmon","submitted_at":"2024-12-10T18:49:39Z","abstract_excerpt":"Text-to-video diffusion models have shown remarkable progress in generating coherent video clips from textual descriptions. However, the interplay between motion, structure, and identity representations in these models remains under-explored. Here, we investigate how self-attention query (Q) features simultaneously govern motion, structure, and identity and examine the challenges arising when these representations interact. Our analysis reveals that Q affects not only layout, but that during denoising Q also has a strong effect on subject identity, making it hard to transfer motion without the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.07750","kind":"arxiv","version":3},"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/2412.07750/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":"2412.07750","created_at":"2026-07-05T11:07:17.748539+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.07750v3","created_at":"2026-07-05T11:07:17.748539+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.07750","created_at":"2026-07-05T11:07:17.748539+00:00"},{"alias_kind":"pith_short_12","alias_value":"RGGC6JDB6VCP","created_at":"2026-07-05T11:07:17.748539+00:00"},{"alias_kind":"pith_short_16","alias_value":"RGGC6JDB6VCPHQUR","created_at":"2026-07-05T11:07:17.748539+00:00"},{"alias_kind":"pith_short_8","alias_value":"RGGC6JDB","created_at":"2026-07-05T11:07:17.748539+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01869","citing_title":"QWERTY: Training-Free Motion Control via Query-Warped Video Diffusion Transformers","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2504.17816","citing_title":"Learning Zero-Shot Subject-Driven Video Generation Using 1% Compute","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22144","citing_title":"One Sentence, One Drama: Personalized Short-Form Drama Generation via Multi-Agent Systems","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6","json":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6.json","graph_json":"https://pith.science/api/pith-number/RGGC6JDB6VCPHQURQQZOQAO5F6/graph.json","events_json":"https://pith.science/api/pith-number/RGGC6JDB6VCPHQURQQZOQAO5F6/events.json","paper":"https://pith.science/paper/RGGC6JDB"},"agent_actions":{"view_html":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6","download_json":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6.json","view_paper":"https://pith.science/paper/RGGC6JDB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.07750&json=true","fetch_graph":"https://pith.science/api/pith-number/RGGC6JDB6VCPHQURQQZOQAO5F6/graph.json","fetch_events":"https://pith.science/api/pith-number/RGGC6JDB6VCPHQURQQZOQAO5F6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6/action/storage_attestation","attest_author":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6/action/author_attestation","sign_citation":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6/action/citation_signature","submit_replication":"https://pith.science/pith/RGGC6JDB6VCPHQURQQZOQAO5F6/action/replication_record"}},"created_at":"2026-07-05T11:07:17.748539+00:00","updated_at":"2026-07-05T11:07:17.748539+00:00"}