{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ECQ5N7TP3PAC3WYNBBYCM7UMXQ","short_pith_number":"pith:ECQ5N7TP","schema_version":"1.0","canonical_sha256":"20a1d6fe6fdbc02ddb0d0870267e8cbc2a27908a3187ed5168cca1e2d2aea91c","source":{"kind":"arxiv","id":"2505.08175","version":3},"attestation_state":"computed","paper":{"title":"Fast Text-to-Audio Generation with Adversarial Post-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Adnan Al-Sinan, CJ Carr, Gian Marco Iodice, Jordi Pons, Josiah Taylor, Julian McAuley, Julian Parker, Taylor Berg-Kirkpatrick, Zachary Novack, Zach Evans, Zack Zukowski","submitted_at":"2025-05-13T02:25:47Z","abstract_excerpt":"Text-to-audio systems, while increasingly performant, are slow at inference time, thus making their latency unpractical for many creative applications. We present Adversarial Relativistic-Contrastive (ARC) post-training, the first adversarial acceleration algorithm for diffusion/flow models not based on distillation. While past adversarial post-training methods have struggled to compare against their expensive distillation counterparts, ARC post-training is a simple procedure that (1) extends a recent relativistic adversarial formulation to diffusion/flow post-training and (2) combines it with"},"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":"2505.08175","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-05-13T02:25:47Z","cross_cats_sorted":["cs.AI","cs.LG","cs.MM","eess.AS"],"title_canon_sha256":"2d9c840d1fd68c041b5e54987140c1952ad8b349abd489f3e4a45c0227b19148","abstract_canon_sha256":"5a35460d1c2ce909f9e44463423c4cd071edeaa9cb072a11b94210eedcfd761e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:44.209483Z","signature_b64":"MLjWKtUzOdbYD+gv8Trkjh9OMpjqh2YbepTYrBbQBw40zwuuipYMBMh0o99eDbOXiuzG2gTyYzCsLqrEfHjACA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20a1d6fe6fdbc02ddb0d0870267e8cbc2a27908a3187ed5168cca1e2d2aea91c","last_reissued_at":"2026-07-05T11:05:44.208985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:44.208985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast Text-to-Audio Generation with Adversarial Post-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Adnan Al-Sinan, CJ Carr, Gian Marco Iodice, Jordi Pons, Josiah Taylor, Julian McAuley, Julian Parker, Taylor Berg-Kirkpatrick, Zachary Novack, Zach Evans, Zack Zukowski","submitted_at":"2025-05-13T02:25:47Z","abstract_excerpt":"Text-to-audio systems, while increasingly performant, are slow at inference time, thus making their latency unpractical for many creative applications. We present Adversarial Relativistic-Contrastive (ARC) post-training, the first adversarial acceleration algorithm for diffusion/flow models not based on distillation. While past adversarial post-training methods have struggled to compare against their expensive distillation counterparts, ARC post-training is a simple procedure that (1) extends a recent relativistic adversarial formulation to diffusion/flow post-training and (2) combines it with"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08175","kind":"arxiv","version":3},"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/2505.08175/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":"2505.08175","created_at":"2026-07-05T11:05:44.209053+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08175v3","created_at":"2026-07-05T11:05:44.209053+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08175","created_at":"2026-07-05T11:05:44.209053+00:00"},{"alias_kind":"pith_short_12","alias_value":"ECQ5N7TP3PAC","created_at":"2026-07-05T11:05:44.209053+00:00"},{"alias_kind":"pith_short_16","alias_value":"ECQ5N7TP3PAC3WYN","created_at":"2026-07-05T11:05:44.209053+00:00"},{"alias_kind":"pith_short_8","alias_value":"ECQ5N7TP","created_at":"2026-07-05T11:05:44.209053+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.10534","citing_title":"WildFX: A DAW-Powered Pipeline for In-the-Wild Audio FX Graph Modeling","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ","json":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ.json","graph_json":"https://pith.science/api/pith-number/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/graph.json","events_json":"https://pith.science/api/pith-number/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/events.json","paper":"https://pith.science/paper/ECQ5N7TP"},"agent_actions":{"view_html":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ","download_json":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ.json","view_paper":"https://pith.science/paper/ECQ5N7TP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08175&json=true","fetch_graph":"https://pith.science/api/pith-number/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/graph.json","fetch_events":"https://pith.science/api/pith-number/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/action/storage_attestation","attest_author":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/action/author_attestation","sign_citation":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/action/citation_signature","submit_replication":"https://pith.science/pith/ECQ5N7TP3PAC3WYNBBYCM7UMXQ/action/replication_record"}},"created_at":"2026-07-05T11:05:44.209053+00:00","updated_at":"2026-07-05T11:05:44.209053+00:00"}