{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EN6EXXZM2EBOWYRDBUTQHRK2LS","short_pith_number":"pith:EN6EXXZM","schema_version":"1.0","canonical_sha256":"237c4bdf2cd102eb62230d2703c55a5c8f61f571cd76596d62d1dc7efcd448f9","source":{"kind":"arxiv","id":"2503.19904","version":1},"attestation_state":"computed","paper":{"title":"Tracktention: Leveraging Point Tracking to Attend Videos Faster and Better","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Andrea Vedaldi, Zihang Lai","submitted_at":"2025-03-25T17:58:48Z","abstract_excerpt":"Temporal consistency is critical in video prediction to ensure that outputs are coherent and free of artifacts. Traditional methods, such as temporal attention and 3D convolution, may struggle with significant object motion and may not capture long-range temporal dependencies in dynamic scenes. To address this gap, we propose the Tracktention Layer, a novel architectural component that explicitly integrates motion information using point tracks, i.e., sequences of corresponding points across frames. By incorporating these motion cues, the Tracktention Layer enhances temporal alignment and effe"},"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":"2503.19904","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-25T17:58:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"28a3def3dd5aa8f54e8539b6e6dcde9ac4bf3cd3d3dcd22454b441479758bdd6","abstract_canon_sha256":"448f095bdf873c80902394980569c1d5cf142d25e3f9837d7f739f460a7da430"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:03.350092Z","signature_b64":"3JosidfTV4Zko+zFiBQKeQr6JcxZmb8xcduabad/lEAuVXvRWxdb8A9LoknT6GOTVcHpwWZJn4z/ybILzXCyAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"237c4bdf2cd102eb62230d2703c55a5c8f61f571cd76596d62d1dc7efcd448f9","last_reissued_at":"2026-07-05T10:39:03.349622Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:03.349622Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tracktention: Leveraging Point Tracking to Attend Videos Faster and Better","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Andrea Vedaldi, Zihang Lai","submitted_at":"2025-03-25T17:58:48Z","abstract_excerpt":"Temporal consistency is critical in video prediction to ensure that outputs are coherent and free of artifacts. Traditional methods, such as temporal attention and 3D convolution, may struggle with significant object motion and may not capture long-range temporal dependencies in dynamic scenes. To address this gap, we propose the Tracktention Layer, a novel architectural component that explicitly integrates motion information using point tracks, i.e., sequences of corresponding points across frames. By incorporating these motion cues, the Tracktention Layer enhances temporal alignment and effe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19904","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/2503.19904/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":"2503.19904","created_at":"2026-07-05T10:39:03.349680+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.19904v1","created_at":"2026-07-05T10:39:03.349680+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19904","created_at":"2026-07-05T10:39:03.349680+00:00"},{"alias_kind":"pith_short_12","alias_value":"EN6EXXZM2EBO","created_at":"2026-07-05T10:39:03.349680+00:00"},{"alias_kind":"pith_short_16","alias_value":"EN6EXXZM2EBOWYRD","created_at":"2026-07-05T10:39:03.349680+00:00"},{"alias_kind":"pith_short_8","alias_value":"EN6EXXZM","created_at":"2026-07-05T10:39:03.349680+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.08694","citing_title":"MoSiC: Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS","json":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS.json","graph_json":"https://pith.science/api/pith-number/EN6EXXZM2EBOWYRDBUTQHRK2LS/graph.json","events_json":"https://pith.science/api/pith-number/EN6EXXZM2EBOWYRDBUTQHRK2LS/events.json","paper":"https://pith.science/paper/EN6EXXZM"},"agent_actions":{"view_html":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS","download_json":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS.json","view_paper":"https://pith.science/paper/EN6EXXZM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.19904&json=true","fetch_graph":"https://pith.science/api/pith-number/EN6EXXZM2EBOWYRDBUTQHRK2LS/graph.json","fetch_events":"https://pith.science/api/pith-number/EN6EXXZM2EBOWYRDBUTQHRK2LS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS/action/storage_attestation","attest_author":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS/action/author_attestation","sign_citation":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS/action/citation_signature","submit_replication":"https://pith.science/pith/EN6EXXZM2EBOWYRDBUTQHRK2LS/action/replication_record"}},"created_at":"2026-07-05T10:39:03.349680+00:00","updated_at":"2026-07-05T10:39:03.349680+00:00"}