{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QE3C5U4VNX424OVM3NDWYCCLC4","short_pith_number":"pith:QE3C5U4V","schema_version":"1.0","canonical_sha256":"81362ed3956df9ae3aacdb476c084b17399d00c6e63583b172054a505a927586","source":{"kind":"arxiv","id":"2502.11234","version":2},"attestation_state":"computed","paper":{"title":"MaskFlow: Discrete Flows For Flexible and Efficient Long Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bj\\\"orn Ommer, Michael Fuest, Vincent Tao Hu","submitted_at":"2025-02-16T18:59:11Z","abstract_excerpt":"Generating long, high-quality videos remains a challenge due to the complex interplay of spatial and temporal dynamics and hardware limitations. In this work, we introduce MaskFlow, a unified video generation framework that combines discrete representations with flow-matching to enable efficient generation of high-quality long videos. By leveraging a frame-level masking strategy during training, MaskFlow conditions on previously generated unmasked frames to generate videos with lengths ten times beyond that of the training sequences. MaskFlow does so very efficiently by enabling the use of fas"},"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":"2502.11234","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-16T18:59:11Z","cross_cats_sorted":[],"title_canon_sha256":"3fceec302728e3e8f64d71fb556b90856214921b614bdab4093654215b574c7b","abstract_canon_sha256":"8ca69015111e6a1e94c487d6499cbd30c57313babb8cf5dd6b242148539f93ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:29:34.276004Z","signature_b64":"92HXO8AMKPHyp8ActJVCD/7irP56KUGHphWox+2SdXXdMwMArPSN7phMmWWgkoM4alTMnJZW7tM7paO6/12BAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81362ed3956df9ae3aacdb476c084b17399d00c6e63583b172054a505a927586","last_reissued_at":"2026-07-05T10:29:34.275212Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:29:34.275212Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MaskFlow: Discrete Flows For Flexible and Efficient Long Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bj\\\"orn Ommer, Michael Fuest, Vincent Tao Hu","submitted_at":"2025-02-16T18:59:11Z","abstract_excerpt":"Generating long, high-quality videos remains a challenge due to the complex interplay of spatial and temporal dynamics and hardware limitations. In this work, we introduce MaskFlow, a unified video generation framework that combines discrete representations with flow-matching to enable efficient generation of high-quality long videos. By leveraging a frame-level masking strategy during training, MaskFlow conditions on previously generated unmasked frames to generate videos with lengths ten times beyond that of the training sequences. MaskFlow does so very efficiently by enabling the use of fas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.11234","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/2502.11234/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":"2502.11234","created_at":"2026-07-05T10:29:34.275322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.11234v2","created_at":"2026-07-05T10:29:34.275322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.11234","created_at":"2026-07-05T10:29:34.275322+00:00"},{"alias_kind":"pith_short_12","alias_value":"QE3C5U4VNX42","created_at":"2026-07-05T10:29:34.275322+00:00"},{"alias_kind":"pith_short_16","alias_value":"QE3C5U4VNX424OVM","created_at":"2026-07-05T10:29:34.275322+00:00"},{"alias_kind":"pith_short_8","alias_value":"QE3C5U4V","created_at":"2026-07-05T10:29:34.275322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.07982","citing_title":"Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09291","citing_title":"dFlowGRPO: Rate-Aware Policy Optimization for Discrete Flow Models","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06491","citing_title":"Discrete Flow Matching Policy Optimization","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4","json":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4.json","graph_json":"https://pith.science/api/pith-number/QE3C5U4VNX424OVM3NDWYCCLC4/graph.json","events_json":"https://pith.science/api/pith-number/QE3C5U4VNX424OVM3NDWYCCLC4/events.json","paper":"https://pith.science/paper/QE3C5U4V"},"agent_actions":{"view_html":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4","download_json":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4.json","view_paper":"https://pith.science/paper/QE3C5U4V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.11234&json=true","fetch_graph":"https://pith.science/api/pith-number/QE3C5U4VNX424OVM3NDWYCCLC4/graph.json","fetch_events":"https://pith.science/api/pith-number/QE3C5U4VNX424OVM3NDWYCCLC4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4/action/storage_attestation","attest_author":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4/action/author_attestation","sign_citation":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4/action/citation_signature","submit_replication":"https://pith.science/pith/QE3C5U4VNX424OVM3NDWYCCLC4/action/replication_record"}},"created_at":"2026-07-05T10:29:34.275322+00:00","updated_at":"2026-07-05T10:29:34.275322+00:00"}