{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VR37FNSXUQI3NMZOZJC4NULEZO","short_pith_number":"pith:VR37FNSX","schema_version":"1.0","canonical_sha256":"ac77f2b657a411b6b32eca45c6d164cbbbc7ff39f083ca7f301d3f6169ff3875","source":{"kind":"arxiv","id":"2206.11894","version":2},"attestation_state":"computed","paper":{"title":"MaskViT: Masked Visual Pre-Training for Video Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.RO"],"primary_cat":"cs.CV","authors_text":"Agrim Gupta, Jiajun Wu, Li Fei-Fei, Roberto Mart\\'in-Mart\\'in, Stephen Tian, Yunzhi Zhang","submitted_at":"2022-06-23T17:59:33Z","abstract_excerpt":"The ability to predict future visual observations conditioned on past observations and motor commands can enable embodied agents to plan solutions to a variety of tasks in complex environments. This work shows that we can create good video prediction models by pre-training transformers via masked visual modeling. Our approach, named MaskViT, is based on two simple design decisions. First, for memory and training efficiency, we use two types of window attention: spatial and spatiotemporal. Second, during training, we mask a variable percentage of tokens instead of a fixed mask ratio. For infere"},"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":"2206.11894","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-06-23T17:59:33Z","cross_cats_sorted":["cs.LG","cs.RO"],"title_canon_sha256":"76d3ff9d0a7791a744b1ab4fd60b86aa35dd82890961ac3670002e4fdb5f45c6","abstract_canon_sha256":"05d1fedbcdae6bcb4fef2fc47bc2701cd5530ab4eee8e41588997dc1ceae5287"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:46:38.624495Z","signature_b64":"4nnJqqVgMvuFdwlRgaTLVcCmI26kBSQy0VMdQ4eFd5WfU2oRDW4QhvMItxtJrcX8PcMhVFXrwToMD0VW6jUeDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac77f2b657a411b6b32eca45c6d164cbbbc7ff39f083ca7f301d3f6169ff3875","last_reissued_at":"2026-07-05T04:46:38.623981Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:46:38.623981Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MaskViT: Masked Visual Pre-Training for Video Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.RO"],"primary_cat":"cs.CV","authors_text":"Agrim Gupta, Jiajun Wu, Li Fei-Fei, Roberto Mart\\'in-Mart\\'in, Stephen Tian, Yunzhi Zhang","submitted_at":"2022-06-23T17:59:33Z","abstract_excerpt":"The ability to predict future visual observations conditioned on past observations and motor commands can enable embodied agents to plan solutions to a variety of tasks in complex environments. This work shows that we can create good video prediction models by pre-training transformers via masked visual modeling. Our approach, named MaskViT, is based on two simple design decisions. First, for memory and training efficiency, we use two types of window attention: spatial and spatiotemporal. Second, during training, we mask a variable percentage of tokens instead of a fixed mask ratio. For infere"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.11894","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/2206.11894/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":"2206.11894","created_at":"2026-07-05T04:46:38.624039+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.11894v2","created_at":"2026-07-05T04:46:38.624039+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.11894","created_at":"2026-07-05T04:46:38.624039+00:00"},{"alias_kind":"pith_short_12","alias_value":"VR37FNSXUQI3","created_at":"2026-07-05T04:46:38.624039+00:00"},{"alias_kind":"pith_short_16","alias_value":"VR37FNSXUQI3NMZO","created_at":"2026-07-05T04:46:38.624039+00:00"},{"alias_kind":"pith_short_8","alias_value":"VR37FNSX","created_at":"2026-07-05T04:46:38.624039+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30542","citing_title":"Physically Viable World Models: A Case for Query-Conditioned Embodied AI","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2603.14392","citing_title":"WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2308.08089","citing_title":"DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory","ref_index":279,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":127,"is_internal_anchor":false},{"citing_arxiv_id":"2602.21668","citing_title":"Space-Time Forecasting of Dynamic Scenes with Motion-aware Gaussian Grouping","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2312.14125","citing_title":"VideoPoet: A Large Language Model for Zero-Shot Video Generation","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2312.13139","citing_title":"Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulation","ref_index":140,"is_internal_anchor":false},{"citing_arxiv_id":"2402.17177","citing_title":"Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models","ref_index":173,"is_internal_anchor":false},{"citing_arxiv_id":"2410.06158","citing_title":"GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulation","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2210.02303","citing_title":"Imagen Video: High Definition Video Generation with Diffusion Models","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2506.09985","citing_title":"V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO","json":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO.json","graph_json":"https://pith.science/api/pith-number/VR37FNSXUQI3NMZOZJC4NULEZO/graph.json","events_json":"https://pith.science/api/pith-number/VR37FNSXUQI3NMZOZJC4NULEZO/events.json","paper":"https://pith.science/paper/VR37FNSX"},"agent_actions":{"view_html":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO","download_json":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO.json","view_paper":"https://pith.science/paper/VR37FNSX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.11894&json=true","fetch_graph":"https://pith.science/api/pith-number/VR37FNSXUQI3NMZOZJC4NULEZO/graph.json","fetch_events":"https://pith.science/api/pith-number/VR37FNSXUQI3NMZOZJC4NULEZO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO/action/storage_attestation","attest_author":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO/action/author_attestation","sign_citation":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO/action/citation_signature","submit_replication":"https://pith.science/pith/VR37FNSXUQI3NMZOZJC4NULEZO/action/replication_record"}},"created_at":"2026-07-05T04:46:38.624039+00:00","updated_at":"2026-07-05T04:46:38.624039+00:00"}