{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:3WRTPH4CTP5H76TX2I2DJVQHCA","short_pith_number":"pith:3WRTPH4C","schema_version":"1.0","canonical_sha256":"dda3379f829bfa7ffa77d23434d6071038b3f436b9c138bccf94f5d03997fe57","source":{"kind":"arxiv","id":"2103.01950","version":1},"attestation_state":"computed","paper":{"title":"Predicting Video with VQVAE","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"A\\\"aron van den Oord, Ali Razavi, Jacob Walker","submitted_at":"2021-03-02T18:59:10Z","abstract_excerpt":"In recent years, the task of video prediction-forecasting future video given past video frames-has attracted attention in the research community. In this paper we propose a novel approach to this problem with Vector Quantized Variational AutoEncoders (VQ-VAE). With VQ-VAE we compress high-resolution videos into a hierarchical set of multi-scale discrete latent variables. Compared to pixels, this compressed latent space has dramatically reduced dimensionality, allowing us to apply scalable autoregressive generative models to predict video. In contrast to previous work that has largely emphasize"},"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":"2103.01950","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-03-02T18:59:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3c87825261c72960e72f997adae521dccca45e24aa36738a6fd1e4e2b8775097","abstract_canon_sha256":"e717def936fed8d1e8192d423ffee5a8bf94cf184b026638e82995e1290f7034"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:19:51.930585Z","signature_b64":"S02e9mxekb3bVZLqSmgRHXh8EC/i+WLr0WPk/fwSlx3z0hooxCHstBbqVzL806axEIgGpAfeG2yD142/khu9Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dda3379f829bfa7ffa77d23434d6071038b3f436b9c138bccf94f5d03997fe57","last_reissued_at":"2026-07-05T02:19:51.930129Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:19:51.930129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Predicting Video with VQVAE","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"A\\\"aron van den Oord, Ali Razavi, Jacob Walker","submitted_at":"2021-03-02T18:59:10Z","abstract_excerpt":"In recent years, the task of video prediction-forecasting future video given past video frames-has attracted attention in the research community. In this paper we propose a novel approach to this problem with Vector Quantized Variational AutoEncoders (VQ-VAE). With VQ-VAE we compress high-resolution videos into a hierarchical set of multi-scale discrete latent variables. Compared to pixels, this compressed latent space has dramatically reduced dimensionality, allowing us to apply scalable autoregressive generative models to predict video. In contrast to previous work that has largely emphasize"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.01950","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/2103.01950/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":"2103.01950","created_at":"2026-07-05T02:19:51.930185+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.01950v1","created_at":"2026-07-05T02:19:51.930185+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.01950","created_at":"2026-07-05T02:19:51.930185+00:00"},{"alias_kind":"pith_short_12","alias_value":"3WRTPH4CTP5H","created_at":"2026-07-05T02:19:51.930185+00:00"},{"alias_kind":"pith_short_16","alias_value":"3WRTPH4CTP5H76TX","created_at":"2026-07-05T02:19:51.930185+00:00"},{"alias_kind":"pith_short_8","alias_value":"3WRTPH4C","created_at":"2026-07-05T02:19:51.930185+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09156","citing_title":"OmniGen-AR: AutoRegressive Any-to-Image Generation","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02385","citing_title":"How Optimality Structures Sparse Dictionaries: A Theory for Understanding SAE Representations","ref_index":270,"is_internal_anchor":false},{"citing_arxiv_id":"2210.02399","citing_title":"Phenaki: Variable Length Video Generation From Open Domain Textual Description","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2310.06114","citing_title":"Learning Interactive Real-World Simulators","ref_index":122,"is_internal_anchor":false},{"citing_arxiv_id":"2211.13221","citing_title":"Latent Video Diffusion Models for High-Fidelity Long Video Generation","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2104.10157","citing_title":"VideoGPT: Video Generation using VQ-VAE and Transformers","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2204.03458","citing_title":"Video Diffusion Models","ref_index":57,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA","json":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA.json","graph_json":"https://pith.science/api/pith-number/3WRTPH4CTP5H76TX2I2DJVQHCA/graph.json","events_json":"https://pith.science/api/pith-number/3WRTPH4CTP5H76TX2I2DJVQHCA/events.json","paper":"https://pith.science/paper/3WRTPH4C"},"agent_actions":{"view_html":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA","download_json":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA.json","view_paper":"https://pith.science/paper/3WRTPH4C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.01950&json=true","fetch_graph":"https://pith.science/api/pith-number/3WRTPH4CTP5H76TX2I2DJVQHCA/graph.json","fetch_events":"https://pith.science/api/pith-number/3WRTPH4CTP5H76TX2I2DJVQHCA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA/action/storage_attestation","attest_author":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA/action/author_attestation","sign_citation":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA/action/citation_signature","submit_replication":"https://pith.science/pith/3WRTPH4CTP5H76TX2I2DJVQHCA/action/replication_record"}},"created_at":"2026-07-05T02:19:51.930185+00:00","updated_at":"2026-07-05T02:19:51.930185+00:00"}