{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:7ZLIUHZYXGSMDHE3SKT75JEDSF","short_pith_number":"pith:7ZLIUHZY","schema_version":"1.0","canonical_sha256":"fe568a1f38b9a4c19c9b92a7fea483917091c075f2f876ff55419db47188823e","source":{"kind":"arxiv","id":"2208.05003","version":1},"attestation_state":"computed","paper":{"title":"Wavelet Score-Based Generative Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Florentin Guth, Simon Coste, Stephane Mallat, Valentin De Bortoli","submitted_at":"2022-08-09T19:13:34Z","abstract_excerpt":"Score-based generative models (SGMs) synthesize new data samples from Gaussian white noise by running a time-reversed Stochastic Differential Equation (SDE) whose drift coefficient depends on some probabilistic score. The discretization of such SDEs typically requires a large number of time steps and hence a high computational cost. This is because of ill-conditioning properties of the score that we analyze mathematically. We show that SGMs can be considerably accelerated, by factorizing the data distribution into a product of conditional probabilities of wavelet coefficients across scales. Th"},"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":"2208.05003","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-08-09T19:13:34Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"4810849a84107ae872268ea62034eb33c8a4055518364251d7c02fbd26ab3e23","abstract_canon_sha256":"82f85386a8ebb94bf5be37a6a421f28bb81e1eb8472760049a17a4e95e39cbc0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:47:30.831456Z","signature_b64":"PElgBXIx3oiSWa5cMPo8Wep2x6F1JEBX5oGG7v+HO7fGfRCgDE33qvDQcvZSBzIpvSGtrWr5xVoH9cqz26zaDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe568a1f38b9a4c19c9b92a7fea483917091c075f2f876ff55419db47188823e","last_reissued_at":"2026-07-05T04:47:30.831045Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:47:30.831045Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wavelet Score-Based Generative Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Florentin Guth, Simon Coste, Stephane Mallat, Valentin De Bortoli","submitted_at":"2022-08-09T19:13:34Z","abstract_excerpt":"Score-based generative models (SGMs) synthesize new data samples from Gaussian white noise by running a time-reversed Stochastic Differential Equation (SDE) whose drift coefficient depends on some probabilistic score. The discretization of such SDEs typically requires a large number of time steps and hence a high computational cost. This is because of ill-conditioning properties of the score that we analyze mathematically. We show that SGMs can be considerably accelerated, by factorizing the data distribution into a product of conditional probabilities of wavelet coefficients across scales. Th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.05003","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/2208.05003/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":"2208.05003","created_at":"2026-07-05T04:47:30.831100+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.05003v1","created_at":"2026-07-05T04:47:30.831100+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.05003","created_at":"2026-07-05T04:47:30.831100+00:00"},{"alias_kind":"pith_short_12","alias_value":"7ZLIUHZYXGSM","created_at":"2026-07-05T04:47:30.831100+00:00"},{"alias_kind":"pith_short_16","alias_value":"7ZLIUHZYXGSMDHE3","created_at":"2026-07-05T04:47:30.831100+00:00"},{"alias_kind":"pith_short_8","alias_value":"7ZLIUHZY","created_at":"2026-07-05T04:47:30.831100+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.06999","citing_title":"Likelihood Training of Cascaded Diffusion Models via Hierarchical Volume-preserving Maps","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF","json":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF.json","graph_json":"https://pith.science/api/pith-number/7ZLIUHZYXGSMDHE3SKT75JEDSF/graph.json","events_json":"https://pith.science/api/pith-number/7ZLIUHZYXGSMDHE3SKT75JEDSF/events.json","paper":"https://pith.science/paper/7ZLIUHZY"},"agent_actions":{"view_html":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF","download_json":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF.json","view_paper":"https://pith.science/paper/7ZLIUHZY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.05003&json=true","fetch_graph":"https://pith.science/api/pith-number/7ZLIUHZYXGSMDHE3SKT75JEDSF/graph.json","fetch_events":"https://pith.science/api/pith-number/7ZLIUHZYXGSMDHE3SKT75JEDSF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF/action/storage_attestation","attest_author":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF/action/author_attestation","sign_citation":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF/action/citation_signature","submit_replication":"https://pith.science/pith/7ZLIUHZYXGSMDHE3SKT75JEDSF/action/replication_record"}},"created_at":"2026-07-05T04:47:30.831100+00:00","updated_at":"2026-07-05T04:47:30.831100+00:00"}