{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PWEVWNWNUYV3GLDPMPQFJN7ICW","short_pith_number":"pith:PWEVWNWN","schema_version":"1.0","canonical_sha256":"7d895b36cda62bb32c6f63e054b7e815a8f1af220e6c23d1de26af42a27d89ea","source":{"kind":"arxiv","id":"2404.02905","version":2},"attestation_state":"computed","paper":{"title":"Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bingyue Peng, Keyu Tian, Liwei Wang, Yi Jiang, Zehuan Yuan","submitted_at":"2024-04-03T17:59:53Z","abstract_excerpt":"We present Visual AutoRegressive modeling (VAR), a new generation paradigm that redefines the autoregressive learning on images as coarse-to-fine \"next-scale prediction\" or \"next-resolution prediction\", diverging from the standard raster-scan \"next-token prediction\". This simple, intuitive methodology allows autoregressive (AR) transformers to learn visual distributions fast and generalize well: VAR, for the first time, makes GPT-like AR models surpass diffusion transformers in image generation. On ImageNet 256x256 benchmark, VAR significantly improve AR baseline by improving Frechet inception"},"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":"2404.02905","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CV","submitted_at":"2024-04-03T17:59:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"965e4c5f6a91948618bbd3b0c194e9521ecb55e8acc0411623d3c206cecf77dd","abstract_canon_sha256":"f93f44f5b51f2e115e508af825a904e26d1d06abe5388a233a2bb69fede2ea6e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:25.165299Z","signature_b64":"m+rfDDl2czrf2fd5XdYbRz5NjuZ9gE46Uyhr+ZQgWZLjimSAurj7X9HLx8pJmYuWt5oTJak4AsDB+144o8zJBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d895b36cda62bb32c6f63e054b7e815a8f1af220e6c23d1de26af42a27d89ea","last_reissued_at":"2026-07-05T08:29:25.164804Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:25.164804Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bingyue Peng, Keyu Tian, Liwei Wang, Yi Jiang, Zehuan Yuan","submitted_at":"2024-04-03T17:59:53Z","abstract_excerpt":"We present Visual AutoRegressive modeling (VAR), a new generation paradigm that redefines the autoregressive learning on images as coarse-to-fine \"next-scale prediction\" or \"next-resolution prediction\", diverging from the standard raster-scan \"next-token prediction\". This simple, intuitive methodology allows autoregressive (AR) transformers to learn visual distributions fast and generalize well: VAR, for the first time, makes GPT-like AR models surpass diffusion transformers in image generation. On ImageNet 256x256 benchmark, VAR significantly improve AR baseline by improving Frechet inception"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.02905","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/2404.02905/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":"2404.02905","created_at":"2026-07-05T08:29:25.164861+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.02905v2","created_at":"2026-07-05T08:29:25.164861+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.02905","created_at":"2026-07-05T08:29:25.164861+00:00"},{"alias_kind":"pith_short_12","alias_value":"PWEVWNWNUYV3","created_at":"2026-07-05T08:29:25.164861+00:00"},{"alias_kind":"pith_short_16","alias_value":"PWEVWNWNUYV3GLDP","created_at":"2026-07-05T08:29:25.164861+00:00"},{"alias_kind":"pith_short_8","alias_value":"PWEVWNWN","created_at":"2026-07-05T08:29:25.164861+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":21,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27088","citing_title":"SubdivAR: Autoregressive Next-Scale Prediction for Neural Mesh Subdivision","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09156","citing_title":"OmniGen-AR: AutoRegressive Any-to-Image Generation","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07107","citing_title":"Coarse-to-Control: Action-Token Planning for Vision-Language-Action Models","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31991","citing_title":"Amplifying Membership Signal Through Chained Regeneration","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27696","citing_title":"Structure over Pixels: Learning Variable-Length Visual Programs","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30341","citing_title":"GPIC: A Giant Permissive Image Corpus for Visual Generation","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21798","citing_title":"CG-MLLM: Captioning and Generating 3D content via Multi-modal Large Language Models","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16384","citing_title":"Mutual Enhancement Between Global Tokens and Patch Tokens: From Theory to Practice","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18390","citing_title":"Vision Foundation Models as Generalist Tokenizers for Image Generation","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2412.14169","citing_title":"Autoregressive Video Generation without Vector Quantization","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2410.13848","citing_title":"Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2503.07265","citing_title":"WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08078","citing_title":"Normalizing Trajectory Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11380","citing_title":"TRACE: Temporal Routing with Autoregressive Cross-channel Experts for EEG Representation Learning","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2406.06525","citing_title":"Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13030","citing_title":"Generative Refinement Networks for Visual Synthesis","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08078","citing_title":"Normalizing Trajectory Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19902","citing_title":"MMCORE: MultiModal COnnection with Representation Aligned Latent Embeddings","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19330","citing_title":"Text-To-Speech with Chain-of-Details: modeling temporal dynamics in speech generation","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW","json":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW.json","graph_json":"https://pith.science/api/pith-number/PWEVWNWNUYV3GLDPMPQFJN7ICW/graph.json","events_json":"https://pith.science/api/pith-number/PWEVWNWNUYV3GLDPMPQFJN7ICW/events.json","paper":"https://pith.science/paper/PWEVWNWN"},"agent_actions":{"view_html":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW","download_json":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW.json","view_paper":"https://pith.science/paper/PWEVWNWN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.02905&json=true","fetch_graph":"https://pith.science/api/pith-number/PWEVWNWNUYV3GLDPMPQFJN7ICW/graph.json","fetch_events":"https://pith.science/api/pith-number/PWEVWNWNUYV3GLDPMPQFJN7ICW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW/action/storage_attestation","attest_author":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW/action/author_attestation","sign_citation":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW/action/citation_signature","submit_replication":"https://pith.science/pith/PWEVWNWNUYV3GLDPMPQFJN7ICW/action/replication_record"}},"created_at":"2026-07-05T08:29:25.164861+00:00","updated_at":"2026-07-05T08:29:25.164861+00:00"}