SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism

[2506.01979] SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Distributed, Parallel, and Cluster Computing arXiv:2506.01979 (cs) [Submitted on 16 May 2025 ( v1 ), last revised 14 Apr 2026 (this version, v4)] Title: SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism Authors: Yuhao Shen , Junyi Shen , Quan Kong , Tianyu Liu , Yao Lu , Cong Wang View a PDF of the paper titled SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism, by Yuhao Shen and 4 other authors View PDF HTML (experimental) Abstract: Recently, speculative decoding (SD) has emerged as a promising technique to accelerate LLM inference by employing a small draft model to propose draft tokens in advance, and validating them in parallel with the large target model. However, the existing SD methods still remain fundamentally constrained by their serialized execution, which causes the mutual waiting bubbles between the draft and target models. To address this challenge, we draw inspiration from branch prediction in modern processors and propose a novel framework \textbf{SpecBranch} to unlock branch parallelism in SD. Specifically, we first take an in-depth analysis of the potential of branch parallelism in SD, and recognize that the key challenge lies in the trade-offs between parallelization and token rollback. Based on the analysis, we strategically introduce parallel speculative branches to preemptively hedge against likely rejections. Meanwhile, to enhance parallelism, we jointly orchestrate adaptive draft lengths with a hybrid combination of the implicit draft model confidence and explicit reusing of target model features. Extensive experiments across various models and benchmarks show that SpecBranch achieves over \textbf{1.8}$\times \sim$ \textbf{4.5}$\times$ speedups against the auto-regressive decoding and reduces rollback tokens by $\textbf{50}$\% for poorly aligned models, realizing its applicability for real-world deployments. Comments: The paper has been accepted by ICLR2026 Subjects: Distributed, Parallel, and Cluster Computing (cs.DC) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2506.01979 [cs.DC] (or arXiv:2506.01979v4 [cs.DC] for this version) https://doi.org/10.48550/arXiv.2506.01979 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Yuhao Shen [ view email ] [v1] Fri, 16 May 2025 07:45:05 UTC (7,986 KB) [v2] Tue, 23 Sep 2025 05:29:24 UTC (7,988 KB) [v3] Thu, 9 Apr 2026 07:07:26 UTC (8,013 KB) [v4] Tue, 14 Apr 2026 05:44:55 UTC (7,986 KB) Full-text links: Access Paper: View a PDF of the paper titled SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism, by Yuhao Shen and 4 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.DC < prev | next > new | recent | 2025-06 Change to browse by: cs cs.AI 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

• HighTide: An Agent-Curated Open-Source VLSI Benchmark Suite#259576
• Python in the front, party in the Backline: compiling quantum workloads across CPUs, GPUs, and FPGAs#673488
• WaveFront Decoding: Parallelized Self-Speculative Decoding for Looped Language Models#1048428
• Compiler-Directed Transformation for Higher-Order Stencils#612030
• Leveraging SIMD for Accelerating Large-number Arithmetic#126482