{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IHYGPVJGLY6N5YDSLEPPIJ53Z3","short_pith_number":"pith:IHYGPVJG","schema_version":"1.0","canonical_sha256":"41f067d5265e3cdee072591ef427bbcef61d43bbae4de735b1283ed0d17632c3","source":{"kind":"arxiv","id":"2607.02214","version":1},"attestation_state":"computed","paper":{"title":"Unlocking Speech-Text Compositional Powers: Instruction-Following Speech Language Models without Instruction Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Congrui Du, Kaizhi Qian, Shiyu Chang, Yang Zhang","submitted_at":"2026-07-02T14:22:46Z","abstract_excerpt":"Instruction tuning for speech language models (SLMs) is substantially more challenging than for text-based large language models (LLMs), as it requires learning a new modality and a wide range of speech-specific instructions in addition to those supported by text LLMs. Existing SLM training approaches largely replicate the text LLM training paradigm by synthesizing large-scale speech pre-training and instruction-tuning datasets. However, this strategy is difficult to scale, since speech sequences are significantly longer than text sequences. In this paper, we propose SpeechCombine, an instruct"},"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":"2607.02214","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T14:22:46Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"35bb8561fe378424ff732172b2ff2f45026c7ab8a3f082d52a31222b83a862f5","abstract_canon_sha256":"147142321a9959675cc671dce89b94c2d04c2d80a64343dbf2fb63227a305249"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:45.298020Z","signature_b64":"afEoSdLYB573AsLGFAjEqkhJE6dQ9lyakUpkPQJAhF5Ite0jlwYpssvnYzSKIUzi97Z+qY4+Bcxq3ww6jQc6Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41f067d5265e3cdee072591ef427bbcef61d43bbae4de735b1283ed0d17632c3","last_reissued_at":"2026-07-03T01:17:45.297581Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:45.297581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unlocking Speech-Text Compositional Powers: Instruction-Following Speech Language Models without Instruction Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Congrui Du, Kaizhi Qian, Shiyu Chang, Yang Zhang","submitted_at":"2026-07-02T14:22:46Z","abstract_excerpt":"Instruction tuning for speech language models (SLMs) is substantially more challenging than for text-based large language models (LLMs), as it requires learning a new modality and a wide range of speech-specific instructions in addition to those supported by text LLMs. Existing SLM training approaches largely replicate the text LLM training paradigm by synthesizing large-scale speech pre-training and instruction-tuning datasets. However, this strategy is difficult to scale, since speech sequences are significantly longer than text sequences. In this paper, we propose SpeechCombine, an instruct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02214","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/2607.02214/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":"2607.02214","created_at":"2026-07-03T01:17:45.297648+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02214v1","created_at":"2026-07-03T01:17:45.297648+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02214","created_at":"2026-07-03T01:17:45.297648+00:00"},{"alias_kind":"pith_short_12","alias_value":"IHYGPVJGLY6N","created_at":"2026-07-03T01:17:45.297648+00:00"},{"alias_kind":"pith_short_16","alias_value":"IHYGPVJGLY6N5YDS","created_at":"2026-07-03T01:17:45.297648+00:00"},{"alias_kind":"pith_short_8","alias_value":"IHYGPVJG","created_at":"2026-07-03T01:17:45.297648+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/IHYGPVJGLY6N5YDSLEPPIJ53Z3","json":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3.json","graph_json":"https://pith.science/api/pith-number/IHYGPVJGLY6N5YDSLEPPIJ53Z3/graph.json","events_json":"https://pith.science/api/pith-number/IHYGPVJGLY6N5YDSLEPPIJ53Z3/events.json","paper":"https://pith.science/paper/IHYGPVJG"},"agent_actions":{"view_html":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3","download_json":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3.json","view_paper":"https://pith.science/paper/IHYGPVJG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02214&json=true","fetch_graph":"https://pith.science/api/pith-number/IHYGPVJGLY6N5YDSLEPPIJ53Z3/graph.json","fetch_events":"https://pith.science/api/pith-number/IHYGPVJGLY6N5YDSLEPPIJ53Z3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3/action/storage_attestation","attest_author":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3/action/author_attestation","sign_citation":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3/action/citation_signature","submit_replication":"https://pith.science/pith/IHYGPVJGLY6N5YDSLEPPIJ53Z3/action/replication_record"}},"created_at":"2026-07-03T01:17:45.297648+00:00","updated_at":"2026-07-03T01:17:45.297648+00:00"}