{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LAQC3IRVGI6WTQILMII4IIVXGS","short_pith_number":"pith:LAQC3IRV","schema_version":"1.0","canonical_sha256":"58202da235323d69c10b6211c422b73484ea705b0182461ab4ebebce678bef1a","source":{"kind":"arxiv","id":"2606.12199","version":1},"attestation_state":"computed","paper":{"title":"Which Speech Representation Better Matches Text-Native Reasoning? A Study of Speech-Text Alignment on Frame Rate and Representation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Chimin Chan, Guangyan Zhang, Haohe Liu, Hongzhan Lin, Peiwen Sun, Qiuqiang Kong, Wei Xue, Xinshen Zhang, Xu Tan, Yiming Li, Zhengxi Liu, Zhen Ye, Zheqi Dai","submitted_at":"2026-06-10T15:19:15Z","abstract_excerpt":"Spoken dialogue models typically start from text LLM backbones, yet reasoning often degrades when conditioning on speech instead of text. We attribute part of this modality gap to a temporal-granularity mismatch: speech tokens are temporally redundant and far longer than text under matched semantics, diluting per-token semantic density and weakening text-native reasoning dynamics. We study speech token design as a representation selection problem and sweep frame rates under a frozen LLM backbone with a fixed information rate. To make low frame rates feasible, we introduce factorized FSQ and a "},"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":"2606.12199","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2026-06-10T15:19:15Z","cross_cats_sorted":["cs.CL","cs.SD"],"title_canon_sha256":"70360d03cad159b71824a303a5f776cbe995cdb1a4d7bc5b67f71a83b06afbaa","abstract_canon_sha256":"a7025a163fbeb83fdfaa6f2321a861c90e5bdcc4510d8206130d7bd62e86c76b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:10:53.916454Z","signature_b64":"uf+sgiyF0wf0m/H9csjS7ggncS+lIfR6sx4VP2s/eplTAnnCxBC3dCuKQlvrOjsYo23ADUosgeI5iq1RcF3jDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58202da235323d69c10b6211c422b73484ea705b0182461ab4ebebce678bef1a","last_reissued_at":"2026-06-11T01:10:53.915586Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:10:53.915586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Which Speech Representation Better Matches Text-Native Reasoning? A Study of Speech-Text Alignment on Frame Rate and Representation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Chimin Chan, Guangyan Zhang, Haohe Liu, Hongzhan Lin, Peiwen Sun, Qiuqiang Kong, Wei Xue, Xinshen Zhang, Xu Tan, Yiming Li, Zhengxi Liu, Zhen Ye, Zheqi Dai","submitted_at":"2026-06-10T15:19:15Z","abstract_excerpt":"Spoken dialogue models typically start from text LLM backbones, yet reasoning often degrades when conditioning on speech instead of text. We attribute part of this modality gap to a temporal-granularity mismatch: speech tokens are temporally redundant and far longer than text under matched semantics, diluting per-token semantic density and weakening text-native reasoning dynamics. We study speech token design as a representation selection problem and sweep frame rates under a frozen LLM backbone with a fixed information rate. To make low frame rates feasible, we introduce factorized FSQ and a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.12199","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/2606.12199/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":"2606.12199","created_at":"2026-06-11T01:10:53.915739+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.12199v1","created_at":"2026-06-11T01:10:53.915739+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.12199","created_at":"2026-06-11T01:10:53.915739+00:00"},{"alias_kind":"pith_short_12","alias_value":"LAQC3IRVGI6W","created_at":"2026-06-11T01:10:53.915739+00:00"},{"alias_kind":"pith_short_16","alias_value":"LAQC3IRVGI6WTQIL","created_at":"2026-06-11T01:10:53.915739+00:00"},{"alias_kind":"pith_short_8","alias_value":"LAQC3IRV","created_at":"2026-06-11T01:10:53.915739+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/LAQC3IRVGI6WTQILMII4IIVXGS","json":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS.json","graph_json":"https://pith.science/api/pith-number/LAQC3IRVGI6WTQILMII4IIVXGS/graph.json","events_json":"https://pith.science/api/pith-number/LAQC3IRVGI6WTQILMII4IIVXGS/events.json","paper":"https://pith.science/paper/LAQC3IRV"},"agent_actions":{"view_html":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS","download_json":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS.json","view_paper":"https://pith.science/paper/LAQC3IRV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.12199&json=true","fetch_graph":"https://pith.science/api/pith-number/LAQC3IRVGI6WTQILMII4IIVXGS/graph.json","fetch_events":"https://pith.science/api/pith-number/LAQC3IRVGI6WTQILMII4IIVXGS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS/action/storage_attestation","attest_author":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS/action/author_attestation","sign_citation":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS/action/citation_signature","submit_replication":"https://pith.science/pith/LAQC3IRVGI6WTQILMII4IIVXGS/action/replication_record"}},"created_at":"2026-06-11T01:10:53.915739+00:00","updated_at":"2026-06-11T01:10:53.915739+00:00"}