{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EAIXI3PSQEU6T5G62EUDAFNNOQ","short_pith_number":"pith:EAIXI3PS","schema_version":"1.0","canonical_sha256":"2011746df28129e9f4ded1283015ad74071b4706cf7c9466e906e5175214e215","source":{"kind":"arxiv","id":"2509.00078","version":1},"attestation_state":"computed","paper":{"title":"ChipChat: Low-Latency Cascaded Conversational Agent in MLX","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Han Tran, Huangjie Zheng, Luke Carlson, Navdeep Jaitly, Richard He Bai, Ruixiang Zhang, Tatiana Likhomanenko, Yizhe Zhang, Zakaria Aldeneh, Zijin Gu","submitted_at":"2025-08-26T20:40:24Z","abstract_excerpt":"The emergence of large language models (LLMs) has transformed spoken dialog systems, yet the optimal architecture for real-time on-device voice agents remains an open question. While end-to-end approaches promise theoretical advantages, cascaded systems (CSs) continue to outperform them in language understanding tasks, despite being constrained by sequential processing latency. In this work, we introduce ChipChat, a novel low-latency CS that overcomes traditional bottlenecks through architectural innovations and streaming optimizations. Our system integrates streaming (a) conversational speech"},"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":"2509.00078","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2025-08-26T20:40:24Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SD"],"title_canon_sha256":"75ce1a0027b007b13a1341f47d5de4a6237c08adfb848cb380267b60c4075ad2","abstract_canon_sha256":"23b7c1e91cb1f2aab0a1cb32314599a91002a4ec74c7ec9616b9b5ad313fe4ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:09.154672Z","signature_b64":"xjUZ/nmxE4Hade5E3r1UEZCmtH23dp2ErafmnK/sQl0RBZxy+xC6LtzbOju4+KwXGKut8ccNBgDIVULfvnNoCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2011746df28129e9f4ded1283015ad74071b4706cf7c9466e906e5175214e215","last_reissued_at":"2026-07-05T12:02:09.154154Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:09.154154Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ChipChat: Low-Latency Cascaded Conversational Agent in MLX","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Han Tran, Huangjie Zheng, Luke Carlson, Navdeep Jaitly, Richard He Bai, Ruixiang Zhang, Tatiana Likhomanenko, Yizhe Zhang, Zakaria Aldeneh, Zijin Gu","submitted_at":"2025-08-26T20:40:24Z","abstract_excerpt":"The emergence of large language models (LLMs) has transformed spoken dialog systems, yet the optimal architecture for real-time on-device voice agents remains an open question. While end-to-end approaches promise theoretical advantages, cascaded systems (CSs) continue to outperform them in language understanding tasks, despite being constrained by sequential processing latency. In this work, we introduce ChipChat, a novel low-latency CS that overcomes traditional bottlenecks through architectural innovations and streaming optimizations. Our system integrates streaming (a) conversational speech"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.00078","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/2509.00078/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":"2509.00078","created_at":"2026-07-05T12:02:09.154217+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.00078v1","created_at":"2026-07-05T12:02:09.154217+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.00078","created_at":"2026-07-05T12:02:09.154217+00:00"},{"alias_kind":"pith_short_12","alias_value":"EAIXI3PSQEU6","created_at":"2026-07-05T12:02:09.154217+00:00"},{"alias_kind":"pith_short_16","alias_value":"EAIXI3PSQEU6T5G6","created_at":"2026-07-05T12:02:09.154217+00:00"},{"alias_kind":"pith_short_8","alias_value":"EAIXI3PS","created_at":"2026-07-05T12:02:09.154217+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.13450","citing_title":"Endpoint Anticipation for Low-Latency Spoken Dialogue","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ","json":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ.json","graph_json":"https://pith.science/api/pith-number/EAIXI3PSQEU6T5G62EUDAFNNOQ/graph.json","events_json":"https://pith.science/api/pith-number/EAIXI3PSQEU6T5G62EUDAFNNOQ/events.json","paper":"https://pith.science/paper/EAIXI3PS"},"agent_actions":{"view_html":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ","download_json":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ.json","view_paper":"https://pith.science/paper/EAIXI3PS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.00078&json=true","fetch_graph":"https://pith.science/api/pith-number/EAIXI3PSQEU6T5G62EUDAFNNOQ/graph.json","fetch_events":"https://pith.science/api/pith-number/EAIXI3PSQEU6T5G62EUDAFNNOQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ/action/storage_attestation","attest_author":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ/action/author_attestation","sign_citation":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ/action/citation_signature","submit_replication":"https://pith.science/pith/EAIXI3PSQEU6T5G62EUDAFNNOQ/action/replication_record"}},"created_at":"2026-07-05T12:02:09.154217+00:00","updated_at":"2026-07-05T12:02:09.154217+00:00"}