{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:QMB5HFISMKWJDSZ6L7CVMKFI6L","short_pith_number":"pith:QMB5HFIS","schema_version":"1.0","canonical_sha256":"8303d3951262ac91cb3e5fc55628a8f2d9ef5ad37f6008afe2a4a1dfab2cde58","source":{"kind":"arxiv","id":"2203.13508","version":1},"attestation_state":"computed","paper":{"title":"BDDM: Bilateral Denoising Diffusion Models for Fast and High-Quality Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SD","eess.SP"],"primary_cat":"eess.AS","authors_text":"Dan Su, Dong Yu, Jun Wang, Max W. Y. Lam","submitted_at":"2022-03-25T08:53:12Z","abstract_excerpt":"Diffusion probabilistic models (DPMs) and their extensions have emerged as competitive generative models yet confront challenges of efficient sampling. We propose a new bilateral denoising diffusion model (BDDM) that parameterizes both the forward and reverse processes with a schedule network and a score network, which can train with a novel bilateral modeling objective. We show that the new surrogate objective can achieve a lower bound of the log marginal likelihood tighter than a conventional surrogate. We also find that BDDM allows inheriting pre-trained score network parameters from any DP"},"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":"2203.13508","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-03-25T08:53:12Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SD","eess.SP"],"title_canon_sha256":"1f9e8f35be12ae27a8bb3de77b544f09e41f27296deb0e1733703d8ca84884bf","abstract_canon_sha256":"807e21f0a10ca9d81b7bca5d56be8f4111df33ae1c865c3a677a6a82709738fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:08:32.934928Z","signature_b64":"SIwS5MmA9iin/nw6uPHA19SAsvPM1uIrn/gQ6vUQTqO4T0KH7dlAJt8zn3EWG87en8go1w4kHzh//ZYXkmC7Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8303d3951262ac91cb3e5fc55628a8f2d9ef5ad37f6008afe2a4a1dfab2cde58","last_reissued_at":"2026-07-05T04:08:32.934540Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:08:32.934540Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BDDM: Bilateral Denoising Diffusion Models for Fast and High-Quality Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SD","eess.SP"],"primary_cat":"eess.AS","authors_text":"Dan Su, Dong Yu, Jun Wang, Max W. Y. Lam","submitted_at":"2022-03-25T08:53:12Z","abstract_excerpt":"Diffusion probabilistic models (DPMs) and their extensions have emerged as competitive generative models yet confront challenges of efficient sampling. We propose a new bilateral denoising diffusion model (BDDM) that parameterizes both the forward and reverse processes with a schedule network and a score network, which can train with a novel bilateral modeling objective. We show that the new surrogate objective can achieve a lower bound of the log marginal likelihood tighter than a conventional surrogate. We also find that BDDM allows inheriting pre-trained score network parameters from any DP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.13508","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/2203.13508/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":"2203.13508","created_at":"2026-07-05T04:08:32.934592+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.13508v1","created_at":"2026-07-05T04:08:32.934592+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.13508","created_at":"2026-07-05T04:08:32.934592+00:00"},{"alias_kind":"pith_short_12","alias_value":"QMB5HFISMKWJ","created_at":"2026-07-05T04:08:32.934592+00:00"},{"alias_kind":"pith_short_16","alias_value":"QMB5HFISMKWJDSZ6","created_at":"2026-07-05T04:08:32.934592+00:00"},{"alias_kind":"pith_short_8","alias_value":"QMB5HFIS","created_at":"2026-07-05T04:08:32.934592+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.03295","citing_title":"CPKD: Clinical Prior Knowledge-Constrained Diffusion Models for Surgical Phase Recognition in Endoscopic Submucosal Dissection","ref_index":43,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L","json":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L.json","graph_json":"https://pith.science/api/pith-number/QMB5HFISMKWJDSZ6L7CVMKFI6L/graph.json","events_json":"https://pith.science/api/pith-number/QMB5HFISMKWJDSZ6L7CVMKFI6L/events.json","paper":"https://pith.science/paper/QMB5HFIS"},"agent_actions":{"view_html":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L","download_json":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L.json","view_paper":"https://pith.science/paper/QMB5HFIS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.13508&json=true","fetch_graph":"https://pith.science/api/pith-number/QMB5HFISMKWJDSZ6L7CVMKFI6L/graph.json","fetch_events":"https://pith.science/api/pith-number/QMB5HFISMKWJDSZ6L7CVMKFI6L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L/action/storage_attestation","attest_author":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L/action/author_attestation","sign_citation":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L/action/citation_signature","submit_replication":"https://pith.science/pith/QMB5HFISMKWJDSZ6L7CVMKFI6L/action/replication_record"}},"created_at":"2026-07-05T04:08:32.934592+00:00","updated_at":"2026-07-05T04:08:32.934592+00:00"}