{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FPJNFMHJVKJRNTQVTNHDSA5ZEV","short_pith_number":"pith:FPJNFMHJ","schema_version":"1.0","canonical_sha256":"2bd2d2b0e9aa9316ce159b4e3903b92566661304f0b95caa429c7274c43b7235","source":{"kind":"arxiv","id":"2410.13863","version":1},"attestation_state":"computed","paper":{"title":"Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Chen Sun, Deqing Sun, Kaiming He, Lijie Fan, Michael Rubinstein, Siyang Qin, Tianhong Li, Yonglong Tian, Yuanzhen Li","submitted_at":"2024-10-17T17:59:59Z","abstract_excerpt":"Scaling up autoregressive models in vision has not proven as beneficial as in large language models. In this work, we investigate this scaling problem in the context of text-to-image generation, focusing on two critical factors: whether models use discrete or continuous tokens, and whether tokens are generated in a random or fixed raster order using BERT- or GPT-like transformer architectures. Our empirical results show that, while all models scale effectively in terms of validation loss, their evaluation performance -- measured by FID, GenEval score, and visual quality -- follows different tr"},"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":"2410.13863","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-17T17:59:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"430c4f420266d19a67e99f8c8ac05c9f4baf749aa6b2774880c97a60a68f70d0","abstract_canon_sha256":"c04aaa40184fb92119153f8409d6b7ac730a193d68f98c85e6745f9c4aef8bdb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:22:06.785366Z","signature_b64":"BB9pjhVDCtSga4BObhshYMxmib7DczMt3J/8O8+MrnKvi/n2O3X/CFs9rcWecOYGwd1mDVAfSM9kfSDmlEbmDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2bd2d2b0e9aa9316ce159b4e3903b92566661304f0b95caa429c7274c43b7235","last_reissued_at":"2026-07-05T09:22:06.784868Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:22:06.784868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Chen Sun, Deqing Sun, Kaiming He, Lijie Fan, Michael Rubinstein, Siyang Qin, Tianhong Li, Yonglong Tian, Yuanzhen Li","submitted_at":"2024-10-17T17:59:59Z","abstract_excerpt":"Scaling up autoregressive models in vision has not proven as beneficial as in large language models. In this work, we investigate this scaling problem in the context of text-to-image generation, focusing on two critical factors: whether models use discrete or continuous tokens, and whether tokens are generated in a random or fixed raster order using BERT- or GPT-like transformer architectures. Our empirical results show that, while all models scale effectively in terms of validation loss, their evaluation performance -- measured by FID, GenEval score, and visual quality -- follows different tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13863","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/2410.13863/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":"2410.13863","created_at":"2026-07-05T09:22:06.784925+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.13863v1","created_at":"2026-07-05T09:22:06.784925+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13863","created_at":"2026-07-05T09:22:06.784925+00:00"},{"alias_kind":"pith_short_12","alias_value":"FPJNFMHJVKJR","created_at":"2026-07-05T09:22:06.784925+00:00"},{"alias_kind":"pith_short_16","alias_value":"FPJNFMHJVKJRNTQV","created_at":"2026-07-05T09:22:06.784925+00:00"},{"alias_kind":"pith_short_8","alias_value":"FPJNFMHJ","created_at":"2026-07-05T09:22:06.784925+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.17069","citing_title":"Distilling Specialized Orders for Visual Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2603.02667","citing_title":"Unifying Contrastive and Generative Objectives for Visual Understanding and Text-to-Image Generation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2505.23606","citing_title":"Muddit: Liberating Generation Beyond Text-to-Image with a Unified Discrete Diffusion Model","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2511.19365","citing_title":"DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2503.19325","citing_title":"Long-Context Autoregressive Video Modeling with Next-Frame Prediction","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02493","citing_title":"PixelGen: Improving Pixel Diffusion with Perceptual Supervision","ref_index":4,"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":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02355","citing_title":"From Broad Exploration to Stable Synthesis: Entropy-Guided Optimization for Autoregressive Image Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06966","citing_title":"MAR-GRPO: Stabilized GRPO for AR-diffusion Hybrid Image Generation","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV","json":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV.json","graph_json":"https://pith.science/api/pith-number/FPJNFMHJVKJRNTQVTNHDSA5ZEV/graph.json","events_json":"https://pith.science/api/pith-number/FPJNFMHJVKJRNTQVTNHDSA5ZEV/events.json","paper":"https://pith.science/paper/FPJNFMHJ"},"agent_actions":{"view_html":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV","download_json":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV.json","view_paper":"https://pith.science/paper/FPJNFMHJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.13863&json=true","fetch_graph":"https://pith.science/api/pith-number/FPJNFMHJVKJRNTQVTNHDSA5ZEV/graph.json","fetch_events":"https://pith.science/api/pith-number/FPJNFMHJVKJRNTQVTNHDSA5ZEV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV/action/storage_attestation","attest_author":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV/action/author_attestation","sign_citation":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV/action/citation_signature","submit_replication":"https://pith.science/pith/FPJNFMHJVKJRNTQVTNHDSA5ZEV/action/replication_record"}},"created_at":"2026-07-05T09:22:06.784925+00:00","updated_at":"2026-07-05T09:22:06.784925+00:00"}