{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XJYHRPXOFXQPOPWGCMH2JRC46I","short_pith_number":"pith:XJYHRPXO","schema_version":"1.0","canonical_sha256":"ba7078beee2de0f73ec6130fa4c45cf22ca29111967e088fd444343a80e020da","source":{"kind":"arxiv","id":"2506.06276","version":1},"attestation_state":"computed","paper":{"title":"STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"David Berthelot, Huangjie Zheng, Jiatao Gu, Josh Susskind, Laurent Dinh, Miguel Angel Bautista, Ruixiang Zhang, Shuangfei Zhai, Tianrong Chen, Yuyang Wang","submitted_at":"2025-06-06T17:58:39Z","abstract_excerpt":"We present STARFlow, a scalable generative model based on normalizing flows that achieves strong performance in high-resolution image synthesis. The core of STARFlow is Transformer Autoregressive Flow (TARFlow), which combines the expressive power of normalizing flows with the structured modeling capabilities of Autoregressive Transformers. We first establish the theoretical universality of TARFlow for modeling continuous distributions. Building on this foundation, we introduce several key architectural and algorithmic innovations to significantly enhance scalability: (1) a deep-shallow design"},"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":"2506.06276","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-06T17:58:39Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"fad1cada80c1d7a0a42abd13d9de9956aef9bc3e9ad592bea5116b8c98ad14fc","abstract_canon_sha256":"c17cc1234c80bda0c1abd63a1ec61380106b1d02cdd0bdfb860b290dfd183263"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:23.981908Z","signature_b64":"yeQNwtdeoxyPm+cFlx+lBnNwFujpgN65/4tZMvAAOfhHKalKDC+1P/XsvvGdcVny0PVw2lCesLB6a6Npip2VDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba7078beee2de0f73ec6130fa4c45cf22ca29111967e088fd444343a80e020da","last_reissued_at":"2026-07-05T11:18:23.981424Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:23.981424Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"David Berthelot, Huangjie Zheng, Jiatao Gu, Josh Susskind, Laurent Dinh, Miguel Angel Bautista, Ruixiang Zhang, Shuangfei Zhai, Tianrong Chen, Yuyang Wang","submitted_at":"2025-06-06T17:58:39Z","abstract_excerpt":"We present STARFlow, a scalable generative model based on normalizing flows that achieves strong performance in high-resolution image synthesis. The core of STARFlow is Transformer Autoregressive Flow (TARFlow), which combines the expressive power of normalizing flows with the structured modeling capabilities of Autoregressive Transformers. We first establish the theoretical universality of TARFlow for modeling continuous distributions. Building on this foundation, we introduce several key architectural and algorithmic innovations to significantly enhance scalability: (1) a deep-shallow design"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06276","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/2506.06276/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":"2506.06276","created_at":"2026-07-05T11:18:23.981486+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.06276v1","created_at":"2026-07-05T11:18:23.981486+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06276","created_at":"2026-07-05T11:18:23.981486+00:00"},{"alias_kind":"pith_short_12","alias_value":"XJYHRPXOFXQP","created_at":"2026-07-05T11:18:23.981486+00:00"},{"alias_kind":"pith_short_16","alias_value":"XJYHRPXOFXQPOPWG","created_at":"2026-07-05T11:18:23.981486+00:00"},{"alias_kind":"pith_short_8","alias_value":"XJYHRPXO","created_at":"2026-07-05T11:18:23.981486+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26016","citing_title":"MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26016","citing_title":"MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20041","citing_title":"Normalizing Flows with Iterative Denoising","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08029","citing_title":"STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I","json":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I.json","graph_json":"https://pith.science/api/pith-number/XJYHRPXOFXQPOPWGCMH2JRC46I/graph.json","events_json":"https://pith.science/api/pith-number/XJYHRPXOFXQPOPWGCMH2JRC46I/events.json","paper":"https://pith.science/paper/XJYHRPXO"},"agent_actions":{"view_html":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I","download_json":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I.json","view_paper":"https://pith.science/paper/XJYHRPXO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.06276&json=true","fetch_graph":"https://pith.science/api/pith-number/XJYHRPXOFXQPOPWGCMH2JRC46I/graph.json","fetch_events":"https://pith.science/api/pith-number/XJYHRPXOFXQPOPWGCMH2JRC46I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I/action/storage_attestation","attest_author":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I/action/author_attestation","sign_citation":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I/action/citation_signature","submit_replication":"https://pith.science/pith/XJYHRPXOFXQPOPWGCMH2JRC46I/action/replication_record"}},"created_at":"2026-07-05T11:18:23.981486+00:00","updated_at":"2026-07-05T11:18:23.981486+00:00"}