{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DNYLA6JJPHW6FRUQRW7MXM5ES7","short_pith_number":"pith:DNYLA6JJ","schema_version":"1.0","canonical_sha256":"1b70b0792979ede2c6908dbecbb3a497dea5ba7ba8f1a3ea0e4aaa2661380bf0","source":{"kind":"arxiv","id":"2506.00722","version":1},"attestation_state":"computed","paper":{"title":"Chain-of-Thought Training for Open E2E Spoken Dialogue Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Emiru Tsunoo, Hayato Futami, Jee-weon Jung, Jiatong Shi, Jinchuan Tian, Shinji Watanabe, Siddhant Arora, Yosuke Kashiwagi","submitted_at":"2025-05-31T21:43:37Z","abstract_excerpt":"Unlike traditional cascaded pipelines, end-to-end (E2E) spoken dialogue systems preserve full differentiability and capture non-phonemic information, making them well-suited for modeling spoken interactions. However, existing E2E approaches often require large-scale training data and generates responses lacking semantic coherence. We propose a simple yet effective strategy leveraging a chain-of-thought (CoT) formulation, ensuring that training on conversational data remains closely aligned with the multimodal language model (LM)'s pre-training on speech recognition~(ASR), text-to-speech synthe"},"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":"2506.00722","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-31T21:43:37Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"7bd0fc62455a8928cf201e7e8bd11092653f095c79ed9c799726d1f4b97a4281","abstract_canon_sha256":"2d83df4d717fdfa9c39a4f7ab7258492707815aa5265aea69800e20e5102263e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:39.301737Z","signature_b64":"TeTDMr+OBbjLh9NmMyEQ7C4KLJkewHssMp8k7kEzgrvAUbpOjQMc8S5gl0GqhWb0o1onJfREE+Jq+HBk5VgLCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b70b0792979ede2c6908dbecbb3a497dea5ba7ba8f1a3ea0e4aaa2661380bf0","last_reissued_at":"2026-07-05T11:13:39.301239Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:39.301239Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Chain-of-Thought Training for Open E2E Spoken Dialogue Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Emiru Tsunoo, Hayato Futami, Jee-weon Jung, Jiatong Shi, Jinchuan Tian, Shinji Watanabe, Siddhant Arora, Yosuke Kashiwagi","submitted_at":"2025-05-31T21:43:37Z","abstract_excerpt":"Unlike traditional cascaded pipelines, end-to-end (E2E) spoken dialogue systems preserve full differentiability and capture non-phonemic information, making them well-suited for modeling spoken interactions. However, existing E2E approaches often require large-scale training data and generates responses lacking semantic coherence. We propose a simple yet effective strategy leveraging a chain-of-thought (CoT) formulation, ensuring that training on conversational data remains closely aligned with the multimodal language model (LM)'s pre-training on speech recognition~(ASR), text-to-speech synthe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00722","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/2506.00722/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":"2506.00722","created_at":"2026-07-05T11:13:39.301295+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.00722v1","created_at":"2026-07-05T11:13:39.301295+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00722","created_at":"2026-07-05T11:13:39.301295+00:00"},{"alias_kind":"pith_short_12","alias_value":"DNYLA6JJPHW6","created_at":"2026-07-05T11:13:39.301295+00:00"},{"alias_kind":"pith_short_16","alias_value":"DNYLA6JJPHW6FRUQ","created_at":"2026-07-05T11:13:39.301295+00:00"},{"alias_kind":"pith_short_8","alias_value":"DNYLA6JJ","created_at":"2026-07-05T11:13:39.301295+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.00722","citing_title":"Chain-of-Thought Training for Open E2E Spoken Dialogue Systems","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7","json":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7.json","graph_json":"https://pith.science/api/pith-number/DNYLA6JJPHW6FRUQRW7MXM5ES7/graph.json","events_json":"https://pith.science/api/pith-number/DNYLA6JJPHW6FRUQRW7MXM5ES7/events.json","paper":"https://pith.science/paper/DNYLA6JJ"},"agent_actions":{"view_html":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7","download_json":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7.json","view_paper":"https://pith.science/paper/DNYLA6JJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.00722&json=true","fetch_graph":"https://pith.science/api/pith-number/DNYLA6JJPHW6FRUQRW7MXM5ES7/graph.json","fetch_events":"https://pith.science/api/pith-number/DNYLA6JJPHW6FRUQRW7MXM5ES7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7/action/storage_attestation","attest_author":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7/action/author_attestation","sign_citation":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7/action/citation_signature","submit_replication":"https://pith.science/pith/DNYLA6JJPHW6FRUQRW7MXM5ES7/action/replication_record"}},"created_at":"2026-07-05T11:13:39.301295+00:00","updated_at":"2026-07-05T11:13:39.301295+00:00"}