{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JTDV5LHIX22OYI6VNCBWKBFZM3","short_pith_number":"pith:JTDV5LHI","schema_version":"1.0","canonical_sha256":"4cc75eace8beb4ec23d568836504b966c78dbb76ee9a66cb9121d28834e17b64","source":{"kind":"arxiv","id":"2502.15681","version":2},"attestation_state":"computed","paper":{"title":"One-step Diffusion Models with $f$-Divergence Distribution Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Arash Vahdat, Weili Nie, Yilun Xu","submitted_at":"2025-02-21T18:59:20Z","abstract_excerpt":"Sampling from diffusion models involves a slow iterative process that hinders their practical deployment, especially for interactive applications. To accelerate generation speed, recent approaches distill a multi-step diffusion model into a single-step student generator via variational score distillation, which matches the distribution of samples generated by the student to the teacher's distribution. However, these approaches use the reverse Kullback-Leibler (KL) divergence for distribution matching which is known to be mode seeking. In this paper, we generalize the distribution matching appr"},"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":"2502.15681","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-21T18:59:20Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"235600ce26a6aefa8f2bee818a74ed5fa0438cc91ec824d4851a23074c87882d","abstract_canon_sha256":"4af833bfcb8ed2ff869a1f8dd4627d10907f6842c8a78731b890e0d67fa2376d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:27:25.924825Z","signature_b64":"F2UWcenRPgZdW2Hn35fd/5QfzFenwbFKW+B6SplHqclrjdlCUNLbvzyBmsUvBXk0gA8W1FngV+HdBDCw/xcFBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cc75eace8beb4ec23d568836504b966c78dbb76ee9a66cb9121d28834e17b64","last_reissued_at":"2026-07-05T10:27:25.924303Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:27:25.924303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"One-step Diffusion Models with $f$-Divergence Distribution Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Arash Vahdat, Weili Nie, Yilun Xu","submitted_at":"2025-02-21T18:59:20Z","abstract_excerpt":"Sampling from diffusion models involves a slow iterative process that hinders their practical deployment, especially for interactive applications. To accelerate generation speed, recent approaches distill a multi-step diffusion model into a single-step student generator via variational score distillation, which matches the distribution of samples generated by the student to the teacher's distribution. However, these approaches use the reverse Kullback-Leibler (KL) divergence for distribution matching which is known to be mode seeking. In this paper, we generalize the distribution matching appr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15681","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/2502.15681/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":"2502.15681","created_at":"2026-07-05T10:27:25.924385+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15681v2","created_at":"2026-07-05T10:27:25.924385+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15681","created_at":"2026-07-05T10:27:25.924385+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTDV5LHIX22O","created_at":"2026-07-05T10:27:25.924385+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTDV5LHIX22OYI6V","created_at":"2026-07-05T10:27:25.924385+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTDV5LHI","created_at":"2026-07-05T10:27:25.924385+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22521","citing_title":"Robust Diffusion Models via Divergence-Induced Weighted Denoising","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21982","citing_title":"CoDMD: Copula-aware Distribution Matching Distillation for Fast Video Generation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18478","citing_title":"Data-Forcing Distillation: Restoring Diversity and Fidelity in Few-Step Video Generation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13426","citing_title":"Accelerating Speculative Diffusions via Block Verification","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05254","citing_title":"Flash-WAM: Modality-Aware Distillation for World Action Models","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.32020","citing_title":"Cross-Space Distillation: Teaching One-Step Students with Modern Diffusion Teachers","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29920","citing_title":"Midpoint Generative Models","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2505.16024","citing_title":"Toward Theoretical Insights into Diffusion Trajectory Distillation via Operator Merging","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18204","citing_title":"Forward-Learned Discrete Diffusion: Learning how to noise to denoise faster","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03623","citing_title":"A Few-Step Generative Model on Cumulative Flow Maps","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23632","citing_title":"Hallo-Live: Real-Time Streaming Joint Audio-Video Avatar Generation with Asynchronous Dual-Stream and Human-Centric Preference Distillation","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07020","citing_title":"FlashMol: High-Quality Molecule Generation in as Few as Four Steps","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3","json":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3.json","graph_json":"https://pith.science/api/pith-number/JTDV5LHIX22OYI6VNCBWKBFZM3/graph.json","events_json":"https://pith.science/api/pith-number/JTDV5LHIX22OYI6VNCBWKBFZM3/events.json","paper":"https://pith.science/paper/JTDV5LHI"},"agent_actions":{"view_html":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3","download_json":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3.json","view_paper":"https://pith.science/paper/JTDV5LHI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15681&json=true","fetch_graph":"https://pith.science/api/pith-number/JTDV5LHIX22OYI6VNCBWKBFZM3/graph.json","fetch_events":"https://pith.science/api/pith-number/JTDV5LHIX22OYI6VNCBWKBFZM3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3/action/storage_attestation","attest_author":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3/action/author_attestation","sign_citation":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3/action/citation_signature","submit_replication":"https://pith.science/pith/JTDV5LHIX22OYI6VNCBWKBFZM3/action/replication_record"}},"created_at":"2026-07-05T10:27:25.924385+00:00","updated_at":"2026-07-05T10:27:25.924385+00:00"}