From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding

[2608.14787] From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Cryptography and Security arXiv:2608.14787 (cs) [Submitted on 14 Aug 2026] Title: From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding Authors: Haoxuan Luo , Jameson Sandler , Ferdinando Fioretto View a PDF of the paper titled From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding, by Haoxuan Luo and 2 other authors View PDF HTML (experimental) Abstract: Speculative decoding is a leading technique to reduce the cost of autoregressive generation by using a small drafter to propose several tokens, which are then verified in parallel by a larger target model. Speculative diffusion decoding (SDD) further removes sequential drafting by generating every position in a draft block in parallel with a discrete diffusion model. However, SDD still invokes the target on every block, leaving verification as a potential bottleneck. This paper recognizes that this creates a new control handle: whether to invoke the verifier at all. Thus, we study verifier skipping, a lossy policy that commits a selected draft prefix directly, and ask which confidence signal should schedule it. Interestingly, our study finds that better token predictors need not yield better schedulers: skips require contiguous high-confidence prefixes, while short skips can induce additional drafting rounds. To study this mismatch, we compare raw confidence with learned marginal and conditional survival scores under the same policy, using Strict SDD, lenience, and top-$k$ acceptance as baselines. On HumanEval with DiffuCoder-7B-Instruct and Qwen3-32B, all three confidence signals save $9.6\%$ to $13.5\%$ of verifier calls at the same observed pass@1 as Strict SDD. Surprisingly, raw confidence saves the most; marginal survival has higher positionwise AUROC than raw confidence at most positions, yet neither learned signal dominates online. Our analysis shows that verifier skipping is a useful new lossy axis and, surprisingly, its key challenge is prefix scheduling rather than token prediction alone. Comments: 14 pages, 6 figures Subjects: Cryptography and Security (cs.CR) ; Computation and Language (cs.CL) Cite as: arXiv:2608.14787 [cs.CR] (or arXiv:2608.14787v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2608.14787 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Haoxuan Luo [ view email ] [v1] Fri, 14 Aug 2026 18:00:08 UTC (151 KB) Full-text links: Access Paper: View a PDF of the paper titled From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding, by Haoxuan Luo and 2 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CR < prev | next > new | recent | 2026-08 Change to browse by: cs cs.CL References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: 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

Related documents

• Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training#667998
• LiLiCorr: Lightweight Likelihood Correlation of Parallel Drafts for Speculative Decoding#659095
• LiLiCorr: Lightweight Likelihood Correlation of Parallel Drafts for Speculative Decoding#656959
• StateFlow: Sequence Pipeline Parallelism for Long-Context Modeling with Linear Recurrence#440341
• GPU-Parallelization of Markov Chain Pool Decoding with Unbiased MCMC#799457