{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:75QNW6WFC76ULOXXCYNMSIFX2D","short_pith_number":"pith:75QNW6WF","schema_version":"1.0","canonical_sha256":"ff60db7ac517fd45baf7161ac920b7d0dcf1eaafd0034b6e62aac8fc095df78b","source":{"kind":"arxiv","id":"2401.08740","version":2},"attestation_state":"computed","paper":{"title":"SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Eric Vanden-Eijnden, Mark Goldstein, Michael S. Albergo, Nanye Ma, Nicholas M. Boffi, Saining Xie","submitted_at":"2024-01-16T18:55:25Z","abstract_excerpt":"We present Scalable Interpolant Transformers (SiT), a family of generative models built on the backbone of Diffusion Transformers (DiT). The interpolant framework, which allows for connecting two distributions in a more flexible way than standard diffusion models, makes possible a modular study of various design choices impacting generative models built on dynamical transport: learning in discrete or continuous time, the objective function, the interpolant that connects the distributions, and deterministic or stochastic sampling. By carefully introducing the above ingredients, SiT surpasses Di"},"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":"2401.08740","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-16T18:55:25Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5de1cd04f65e86afdc96e5d941554aee355489601ddc539e475eb36aa8dd1877","abstract_canon_sha256":"2c8c4f97165d19669ab8f23a82d6f6a7259ce4c8254c181f92791530641807d4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:29.980359Z","signature_b64":"S0LvGeQgzRguhHGSt0HMYNOrq8n0fHYMBJCcy0SEI/buwEHUbcWFFUbZUkqG+jQVgqAX3BAZnXCmAOM9M5OWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff60db7ac517fd45baf7161ac920b7d0dcf1eaafd0034b6e62aac8fc095df78b","last_reissued_at":"2026-07-05T09:10:29.979926Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:29.979926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Eric Vanden-Eijnden, Mark Goldstein, Michael S. Albergo, Nanye Ma, Nicholas M. Boffi, Saining Xie","submitted_at":"2024-01-16T18:55:25Z","abstract_excerpt":"We present Scalable Interpolant Transformers (SiT), a family of generative models built on the backbone of Diffusion Transformers (DiT). The interpolant framework, which allows for connecting two distributions in a more flexible way than standard diffusion models, makes possible a modular study of various design choices impacting generative models built on dynamical transport: learning in discrete or continuous time, the objective function, the interpolant that connects the distributions, and deterministic or stochastic sampling. By carefully introducing the above ingredients, SiT surpasses Di"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.08740","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/2401.08740/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":"2401.08740","created_at":"2026-07-05T09:10:29.979982+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.08740v2","created_at":"2026-07-05T09:10:29.979982+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.08740","created_at":"2026-07-05T09:10:29.979982+00:00"},{"alias_kind":"pith_short_12","alias_value":"75QNW6WFC76U","created_at":"2026-07-05T09:10:29.979982+00:00"},{"alias_kind":"pith_short_16","alias_value":"75QNW6WFC76ULOXX","created_at":"2026-07-05T09:10:29.979982+00:00"},{"alias_kind":"pith_short_8","alias_value":"75QNW6WF","created_at":"2026-07-05T09:10:29.979982+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":25,"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":26,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11096","citing_title":"IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05552","citing_title":"Balancing Image Compression and Generation with Bootstrapped Tokenization","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04432","citing_title":"DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04165","citing_title":"CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters","ref_index":91,"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":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27760","citing_title":"PixelU: A U-Shaped Transformer for Efficient End-to-End Pixel Diffusion","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2410.07430","citing_title":"EventFlow: Forecasting Temporal Point Processes with Flow Matching","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17759","citing_title":"FrequencyBooster: Full-Frequency Modeling for High-Fidelity Pixel Diffusion","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":27,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09732","citing_title":"Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2509.13576","citing_title":"Cross-Distribution Diffusion Priors-Driven Iterative Reconstruction for Sparse-View CT","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2509.18611","citing_title":"Flow marching for a generative PDE foundation model","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2511.19365","citing_title":"DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02493","citing_title":"PixelGen: Improving Pixel Diffusion with Perceptual Supervision","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2603.14186","citing_title":"Setting-Matched and Semantics-Scaled Benchmarking of One-Step Generative Models Against Multistep Diffusion and Flow Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26571","citing_title":"GVCC: Zero-Shot Video Compression via Codebook-Driven Stochastic Rectified Flow","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2504.11346","citing_title":"Seedream 3.0 Technical Report","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06264","citing_title":"Flow Matching Guide and Code","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08344","citing_title":"What Time Is It? How Data Geometry Makes Time Conditioning Optional for Flow Matching","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10019","citing_title":"The two clocks and the innovation window: When and how generative models learn rules","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00825","citing_title":"Posterior Augmented Flow Matching","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08837","citing_title":"Discrete Meanflow Training Curriculum","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19141","citing_title":"Denoising, Fast and Slow: Difficulty-Aware Adaptive Sampling for Image Generation","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D","json":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D.json","graph_json":"https://pith.science/api/pith-number/75QNW6WFC76ULOXXCYNMSIFX2D/graph.json","events_json":"https://pith.science/api/pith-number/75QNW6WFC76ULOXXCYNMSIFX2D/events.json","paper":"https://pith.science/paper/75QNW6WF"},"agent_actions":{"view_html":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D","download_json":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D.json","view_paper":"https://pith.science/paper/75QNW6WF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.08740&json=true","fetch_graph":"https://pith.science/api/pith-number/75QNW6WFC76ULOXXCYNMSIFX2D/graph.json","fetch_events":"https://pith.science/api/pith-number/75QNW6WFC76ULOXXCYNMSIFX2D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D/action/storage_attestation","attest_author":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D/action/author_attestation","sign_citation":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D/action/citation_signature","submit_replication":"https://pith.science/pith/75QNW6WFC76ULOXXCYNMSIFX2D/action/replication_record"}},"created_at":"2026-07-05T09:10:29.979982+00:00","updated_at":"2026-07-05T09:10:29.979982+00:00"}