SPAR-K: Scheduled Periodic Alternating Early Exit for Spoken Language Models

[2603.09215] SPAR-K: Scheduled Periodic Alternating Early Exit for Spoken Language Models Skip to main content Search Submit Donate Log in Computer Science > Computation and Language arXiv:2603.09215 (cs) [Submitted on 10 Mar 2026 ( v1 ), last revised 27 Aug 2026 (this version, v2)] Title: SPAR-K: Scheduled Periodic Alternating Early Exit for Spoken Language Models Authors: Hsiao-Ying Huang , Cheng-Han Chiang , Hung-yi Lee View PDF HTML (experimental) Abstract: Interleaved spoken language models (SLMs) alternately generate text and speech tokens, but decoding at full transformer depth for every step becomes costly, especially due to long speech sequences. We propose SPAR-K, a modality-aware early exit framework designed to accelerate interleaved SLM inference while preserving perceptual quality. SPAR-K introduces a speech alternating-depth schedule: most speech positions exit at a fixed intermediate layer, while periodic full-depth "refresh" steps mitigate distribution shift due to early exit. We evaluate our framework using Step-Audio-2-mini and GLM-4-Voice across four datasets spanning reasoning, factual QA, and dialogue tasks, measuring performance in terms of ASR transcription accuracy and perceptual quality. Experimental results demonstrate that SPAR-K largely preserves question-answering accuracy with a maximum accuracy drop of 0.82\% while reducing average speech decoding depth by up to 11\% on Step-Audio-2-mini and 5\% on GLM-4-Voice, both with negligible changes in MOS and WER and no auxiliary computation overhead. We further demonstrate that confidence-based early exit strategies, widely used in text LLMs, are suboptimal for SLMs, highlighting that the unique statistical nature of speech tokens necessitates a specialized early exit design. Comments: 8 pages, 1 figure, 2 tables Subjects: Computation and Language (cs.CL) ; Audio and Speech Processing (eess.AS) Cite as: arXiv:2603.09215 [cs.CL] (or arXiv:2603.09215v2 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2603.09215 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Hsiao-Ying Huang [ view email ] [v1] Tue, 10 Mar 2026 05:39:03 UTC (1,387 KB) [v2] Thu, 27 Aug 2026 08:44:51 UTC (1,446 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.CL < prev | next > new | recent | 2026-03 Change to browse by: cs eess eess.AS References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from (opens in new tab) (opens in new tab) (opens in new tab)

Related documents

• SPAR-K: Scheduled Periodic Alternating Early Exit for Spoken Language Models#677518
• Learning Natural Conversational Behavior in Tandem Speech-to-Speech Models with Randomized Guidance#1105332
• Entropy Bandwidth Y-Axis (EBY)#705619
• Elastic Selective Spectral Hybrids for Train-Once, Export-Many Budgeted Inference#1116817
• Spatio-temporal analysis of spontaneous speech with microphone arrays#401898