{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5U5JQGEBDRK3LU2GEU5ZN3UN2R","short_pith_number":"pith:5U5JQGEB","schema_version":"1.0","canonical_sha256":"ed3a9818811c55b5d346253b96ee8dd44db6a2cb55bb5fc55764c12041ab1323","source":{"kind":"arxiv","id":"2402.04650","version":4},"attestation_state":"computed","paper":{"title":"An analysis of the noise schedule for score-based generative models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Antonio Ocello (CMAP), Claire Boyer (LPSM (UMR\\_8001), IUF), LPSM (UMR\\_8001)), Stanislas Strasman (SU, SU), Sylvain Le Corff (LPSM (UMR\\_8001), Vincent Lemaire (LPSM (UMR\\_8001))","submitted_at":"2024-02-07T08:24:35Z","abstract_excerpt":"Score-based generative models (SGMs) aim at estimating a target data distribution by learning score functions using only noise-perturbed samples from the target.Recent literature has focused extensively on assessing the error between the target and estimated distributions, gauging the generative quality through the Kullback-Leibler (KL) divergence and Wasserstein distances.  Under mild assumptions on the data distribution, we establish an upper bound for the KL divergence between the target and the estimated distributions, explicitly depending on any time-dependent noise schedule. Under additi"},"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":"2402.04650","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2024-02-07T08:24:35Z","cross_cats_sorted":["stat.ML","stat.TH"],"title_canon_sha256":"b9a9380ed0028049c4ad926a18fd1cd872908e83b0185b8d7ddff03b9a80fea4","abstract_canon_sha256":"57b1f63ce7bc308eea047b853522d01d7705bdf3858af4dece7bd15f8c78983a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:05:41.489038Z","signature_b64":"Dcf7WqecDEhny5GqaKapXZ/qNUqXbjHtA67SM3eegReuywKtKhVftqMD3ZMi1viFbaKYiVzgvUEJhAOzkkoaDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed3a9818811c55b5d346253b96ee8dd44db6a2cb55bb5fc55764c12041ab1323","last_reissued_at":"2026-07-05T10:05:41.488525Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:05:41.488525Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An analysis of the noise schedule for score-based generative models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Antonio Ocello (CMAP), Claire Boyer (LPSM (UMR\\_8001), IUF), LPSM (UMR\\_8001)), Stanislas Strasman (SU, SU), Sylvain Le Corff (LPSM (UMR\\_8001), Vincent Lemaire (LPSM (UMR\\_8001))","submitted_at":"2024-02-07T08:24:35Z","abstract_excerpt":"Score-based generative models (SGMs) aim at estimating a target data distribution by learning score functions using only noise-perturbed samples from the target.Recent literature has focused extensively on assessing the error between the target and estimated distributions, gauging the generative quality through the Kullback-Leibler (KL) divergence and Wasserstein distances.  Under mild assumptions on the data distribution, we establish an upper bound for the KL divergence between the target and the estimated distributions, explicitly depending on any time-dependent noise schedule. Under additi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.04650","kind":"arxiv","version":4},"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/2402.04650/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":"2402.04650","created_at":"2026-07-05T10:05:41.488587+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.04650v4","created_at":"2026-07-05T10:05:41.488587+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.04650","created_at":"2026-07-05T10:05:41.488587+00:00"},{"alias_kind":"pith_short_12","alias_value":"5U5JQGEBDRK3","created_at":"2026-07-05T10:05:41.488587+00:00"},{"alias_kind":"pith_short_16","alias_value":"5U5JQGEBDRK3LU2G","created_at":"2026-07-05T10:05:41.488587+00:00"},{"alias_kind":"pith_short_8","alias_value":"5U5JQGEB","created_at":"2026-07-05T10:05:41.488587+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14276","citing_title":"Training-Free Generative Sampling via Moment-Matched Score Smoothing","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06065","citing_title":"Lipschitz regularity in Flow Matching and Diffusion Models: sharp sampling rates and functional inequalities","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R","json":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R.json","graph_json":"https://pith.science/api/pith-number/5U5JQGEBDRK3LU2GEU5ZN3UN2R/graph.json","events_json":"https://pith.science/api/pith-number/5U5JQGEBDRK3LU2GEU5ZN3UN2R/events.json","paper":"https://pith.science/paper/5U5JQGEB"},"agent_actions":{"view_html":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R","download_json":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R.json","view_paper":"https://pith.science/paper/5U5JQGEB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.04650&json=true","fetch_graph":"https://pith.science/api/pith-number/5U5JQGEBDRK3LU2GEU5ZN3UN2R/graph.json","fetch_events":"https://pith.science/api/pith-number/5U5JQGEBDRK3LU2GEU5ZN3UN2R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R/action/storage_attestation","attest_author":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R/action/author_attestation","sign_citation":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R/action/citation_signature","submit_replication":"https://pith.science/pith/5U5JQGEBDRK3LU2GEU5ZN3UN2R/action/replication_record"}},"created_at":"2026-07-05T10:05:41.488587+00:00","updated_at":"2026-07-05T10:05:41.488587+00:00"}