{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:E64ORHKDHTNNJDAPIK7CT5APQG","short_pith_number":"pith:E64ORHKD","schema_version":"1.0","canonical_sha256":"27b8e89d433cdad48c0f42be29f40f81977068367f1e88e2af1fbc79eddda5af","source":{"kind":"arxiv","id":"2212.05199","version":2},"attestation_state":"computed","paper":{"title":"MAGVIT: Masked Generative Video Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexander G. Hauptmann, Han Zhang, Huiwen Chang, Irfan Essa, Jos\\'e Lezama, Kihyuk Sohn, Lijun Yu, Lu Jiang, Ming-Hsuan Yang, Yong Cheng, Yuan Hao","submitted_at":"2022-12-10T04:26:32Z","abstract_excerpt":"We introduce the MAsked Generative VIdeo Transformer, MAGVIT, to tackle various video synthesis tasks with a single model. We introduce a 3D tokenizer to quantize a video into spatial-temporal visual tokens and propose an embedding method for masked video token modeling to facilitate multi-task learning. We conduct extensive experiments to demonstrate the quality, efficiency, and flexibility of MAGVIT. Our experiments show that (i) MAGVIT performs favorably against state-of-the-art approaches and establishes the best-published FVD on three video generation benchmarks, including the challenging"},"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":"2212.05199","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-12-10T04:26:32Z","cross_cats_sorted":[],"title_canon_sha256":"95bfe1bd67af25d5cf8f9f265cf1a9cff0027442da8f0bc006699ad58925d3cf","abstract_canon_sha256":"5c9ed04b20fa26f1d1cdc075fce9e4dbe02e39141f5fee931c614ff0a0c175ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:58:18.601657Z","signature_b64":"Z8bCmefCRkH4kY9RG0OTnPGfBz9m5sY4BUnMDM9Ob8DHnaoRAnSUY/7t1jPzuPaZPbSPhGb6UzKRJPHfZMUWCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"27b8e89d433cdad48c0f42be29f40f81977068367f1e88e2af1fbc79eddda5af","last_reissued_at":"2026-07-05T05:58:18.601100Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:58:18.601100Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAGVIT: Masked Generative Video Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexander G. Hauptmann, Han Zhang, Huiwen Chang, Irfan Essa, Jos\\'e Lezama, Kihyuk Sohn, Lijun Yu, Lu Jiang, Ming-Hsuan Yang, Yong Cheng, Yuan Hao","submitted_at":"2022-12-10T04:26:32Z","abstract_excerpt":"We introduce the MAsked Generative VIdeo Transformer, MAGVIT, to tackle various video synthesis tasks with a single model. We introduce a 3D tokenizer to quantize a video into spatial-temporal visual tokens and propose an embedding method for masked video token modeling to facilitate multi-task learning. We conduct extensive experiments to demonstrate the quality, efficiency, and flexibility of MAGVIT. Our experiments show that (i) MAGVIT performs favorably against state-of-the-art approaches and establishes the best-published FVD on three video generation benchmarks, including the challenging"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.05199","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/2212.05199/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":"2212.05199","created_at":"2026-07-05T05:58:18.601170+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.05199v2","created_at":"2026-07-05T05:58:18.601170+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.05199","created_at":"2026-07-05T05:58:18.601170+00:00"},{"alias_kind":"pith_short_12","alias_value":"E64ORHKDHTNN","created_at":"2026-07-05T05:58:18.601170+00:00"},{"alias_kind":"pith_short_16","alias_value":"E64ORHKDHTNNJDAP","created_at":"2026-07-05T05:58:18.601170+00:00"},{"alias_kind":"pith_short_8","alias_value":"E64ORHKD","created_at":"2026-07-05T05:58:18.601170+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02631","citing_title":"Wavelet as Tokenizer: Preliminary Results on a Shared Wavelet Token Schema for Natural Signals","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08674","citing_title":"BioVid: Autoregressive Video Generation with Biological Behavior Semantic Comprehension","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12090","citing_title":"World Action Models: The Next Frontier in Embodied AI","ref_index":288,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09747","citing_title":"FAST: Efficient Action Tokenization for Vision-Language-Action Models","ref_index":67,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG","json":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG.json","graph_json":"https://pith.science/api/pith-number/E64ORHKDHTNNJDAPIK7CT5APQG/graph.json","events_json":"https://pith.science/api/pith-number/E64ORHKDHTNNJDAPIK7CT5APQG/events.json","paper":"https://pith.science/paper/E64ORHKD"},"agent_actions":{"view_html":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG","download_json":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG.json","view_paper":"https://pith.science/paper/E64ORHKD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.05199&json=true","fetch_graph":"https://pith.science/api/pith-number/E64ORHKDHTNNJDAPIK7CT5APQG/graph.json","fetch_events":"https://pith.science/api/pith-number/E64ORHKDHTNNJDAPIK7CT5APQG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG/action/storage_attestation","attest_author":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG/action/author_attestation","sign_citation":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG/action/citation_signature","submit_replication":"https://pith.science/pith/E64ORHKDHTNNJDAPIK7CT5APQG/action/replication_record"}},"created_at":"2026-07-05T05:58:18.601170+00:00","updated_at":"2026-07-05T05:58:18.601170+00:00"}