{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZMBFAZFHCNT5TPN5RA2E4GBEMP","short_pith_number":"pith:ZMBFAZFH","schema_version":"1.0","canonical_sha256":"cb025064a71367d9bdbd88344e182463da4820743e9e7cef2d037cc3891b8ece","source":{"kind":"arxiv","id":"2508.16306","version":1},"attestation_state":"computed","paper":{"title":"A Sharp KL-Convergence Analysis for Diffusion Models under Minimal Assumptions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.AP","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Nishant Jain, Tong Zhang","submitted_at":"2025-08-22T11:29:06Z","abstract_excerpt":"Diffusion-based generative models have emerged as highly effective methods for synthesizing high-quality samples. Recent works have focused on analyzing the convergence of their generation process with minimal assumptions, either through reverse SDEs or Probability Flow ODEs. The best known guarantees, without any smoothness assumptions, for the KL divergence so far achieve a linear dependence on the data dimension $d$ and an inverse quadratic dependence on $\\varepsilon$. In this work, we present a refined analysis that improves the dependence on $\\varepsilon$. We model the generation process "},"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":"2508.16306","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2025-08-22T11:29:06Z","cross_cats_sorted":["cs.LG","math.AP","math.ST","stat.TH"],"title_canon_sha256":"2f55583c3b9e0fc4331b0a64e8b55c70eba76fcdb8d686460a9dc936acd2d120","abstract_canon_sha256":"6e746f6e48137e6f31a770889c733deda03519c04b061650ba59f03824cb1be0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:45.822522Z","signature_b64":"+axpP4p+sOWuqngs4kHPWwZ3WppWRnVylRgdwtDv0ctWdMtOpH5re3PN66tYi2cg4BqqY/6//CyIrxfhm47SCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb025064a71367d9bdbd88344e182463da4820743e9e7cef2d037cc3891b8ece","last_reissued_at":"2026-07-05T11:57:45.821821Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:45.821821Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Sharp KL-Convergence Analysis for Diffusion Models under Minimal Assumptions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.AP","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Nishant Jain, Tong Zhang","submitted_at":"2025-08-22T11:29:06Z","abstract_excerpt":"Diffusion-based generative models have emerged as highly effective methods for synthesizing high-quality samples. Recent works have focused on analyzing the convergence of their generation process with minimal assumptions, either through reverse SDEs or Probability Flow ODEs. The best known guarantees, without any smoothness assumptions, for the KL divergence so far achieve a linear dependence on the data dimension $d$ and an inverse quadratic dependence on $\\varepsilon$. In this work, we present a refined analysis that improves the dependence on $\\varepsilon$. We model the generation process "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.16306","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/2508.16306/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":"2508.16306","created_at":"2026-07-05T11:57:45.821935+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.16306v1","created_at":"2026-07-05T11:57:45.821935+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.16306","created_at":"2026-07-05T11:57:45.821935+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZMBFAZFHCNT5","created_at":"2026-07-05T11:57:45.821935+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZMBFAZFHCNT5TPN5","created_at":"2026-07-05T11:57:45.821935+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZMBFAZFH","created_at":"2026-07-05T11:57:45.821935+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.22723","citing_title":"The Value of Covariance Matching in Gaussian DDPMs and the Lanczos Sampler","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP","json":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP.json","graph_json":"https://pith.science/api/pith-number/ZMBFAZFHCNT5TPN5RA2E4GBEMP/graph.json","events_json":"https://pith.science/api/pith-number/ZMBFAZFHCNT5TPN5RA2E4GBEMP/events.json","paper":"https://pith.science/paper/ZMBFAZFH"},"agent_actions":{"view_html":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP","download_json":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP.json","view_paper":"https://pith.science/paper/ZMBFAZFH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.16306&json=true","fetch_graph":"https://pith.science/api/pith-number/ZMBFAZFHCNT5TPN5RA2E4GBEMP/graph.json","fetch_events":"https://pith.science/api/pith-number/ZMBFAZFHCNT5TPN5RA2E4GBEMP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP/action/storage_attestation","attest_author":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP/action/author_attestation","sign_citation":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP/action/citation_signature","submit_replication":"https://pith.science/pith/ZMBFAZFHCNT5TPN5RA2E4GBEMP/action/replication_record"}},"created_at":"2026-07-05T11:57:45.821935+00:00","updated_at":"2026-07-05T11:57:45.821935+00:00"}