{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5P2DDVRUBSOQWWTDTE2Z5OHAEY","short_pith_number":"pith:5P2DDVRU","schema_version":"1.0","canonical_sha256":"ebf431d6340c9d0b5a6399359eb8e0263b4e1a5511ea268607647cff1c601da4","source":{"kind":"arxiv","id":"2605.25930","version":1},"attestation_state":"computed","paper":{"title":"CosyEdit2: Speech-Editing-Oriented Reinforcement Learning Unlocks Better Zero-Shot TTS","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Hui Wang, Jiaming Zhou, Junyang Chen, Yongchang Gan, Yong Qin, Yuhang Jia","submitted_at":"2026-05-25T15:12:56Z","abstract_excerpt":"Speech editing and zero-shot Text-to-Speech (TTS) share a similar generative foundation conditioned on speech prompts, yet speech editing demands far stricter local acoustic consistency with surrounding unedited content. While prior work has shown that Supervised Fine-Tuning (SFT) enables TTS models to acquire functional editing capability, this approach remains fundamentally bottlenecked by imperfect paired editing data and coarse-grained optimization signals. To address these limitations, we propose CosyEdit2, a speech editing model built on a two-stage post-training framework that progresse"},"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":"2605.25930","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2026-05-25T15:12:56Z","cross_cats_sorted":[],"title_canon_sha256":"fcfdf26a09795c6657403c3c31aab9cab8c9431be2de532af7c8ef9d669472a0","abstract_canon_sha256":"3a30f5bf152b719177f38ed6a247fb0de5a25c958258b1c712777ecab8e3bc45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:05:18.880071Z","signature_b64":"7+PoD6IMnmEJqWDZLAxFGs4CZ9r+/Ur+VacQK/IvYWeqr96SwuH5EAQBcZ15lL52Zh0An5hkEg97u01LrYKuAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ebf431d6340c9d0b5a6399359eb8e0263b4e1a5511ea268607647cff1c601da4","last_reissued_at":"2026-05-26T02:05:18.879331Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:05:18.879331Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CosyEdit2: Speech-Editing-Oriented Reinforcement Learning Unlocks Better Zero-Shot TTS","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Hui Wang, Jiaming Zhou, Junyang Chen, Yongchang Gan, Yong Qin, Yuhang Jia","submitted_at":"2026-05-25T15:12:56Z","abstract_excerpt":"Speech editing and zero-shot Text-to-Speech (TTS) share a similar generative foundation conditioned on speech prompts, yet speech editing demands far stricter local acoustic consistency with surrounding unedited content. While prior work has shown that Supervised Fine-Tuning (SFT) enables TTS models to acquire functional editing capability, this approach remains fundamentally bottlenecked by imperfect paired editing data and coarse-grained optimization signals. To address these limitations, we propose CosyEdit2, a speech editing model built on a two-stage post-training framework that progresse"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.25930","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/2605.25930/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":"2605.25930","created_at":"2026-05-26T02:05:18.879442+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.25930v1","created_at":"2026-05-26T02:05:18.879442+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.25930","created_at":"2026-05-26T02:05:18.879442+00:00"},{"alias_kind":"pith_short_12","alias_value":"5P2DDVRUBSOQ","created_at":"2026-05-26T02:05:18.879442+00:00"},{"alias_kind":"pith_short_16","alias_value":"5P2DDVRUBSOQWWTD","created_at":"2026-05-26T02:05:18.879442+00:00"},{"alias_kind":"pith_short_8","alias_value":"5P2DDVRU","created_at":"2026-05-26T02:05:18.879442+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY","json":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY.json","graph_json":"https://pith.science/api/pith-number/5P2DDVRUBSOQWWTDTE2Z5OHAEY/graph.json","events_json":"https://pith.science/api/pith-number/5P2DDVRUBSOQWWTDTE2Z5OHAEY/events.json","paper":"https://pith.science/paper/5P2DDVRU"},"agent_actions":{"view_html":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY","download_json":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY.json","view_paper":"https://pith.science/paper/5P2DDVRU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.25930&json=true","fetch_graph":"https://pith.science/api/pith-number/5P2DDVRUBSOQWWTDTE2Z5OHAEY/graph.json","fetch_events":"https://pith.science/api/pith-number/5P2DDVRUBSOQWWTDTE2Z5OHAEY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY/action/storage_attestation","attest_author":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY/action/author_attestation","sign_citation":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY/action/citation_signature","submit_replication":"https://pith.science/pith/5P2DDVRUBSOQWWTDTE2Z5OHAEY/action/replication_record"}},"created_at":"2026-05-26T02:05:18.879442+00:00","updated_at":"2026-05-26T02:05:18.879442+00:00"}