The efficiency of Large Language Model (LLM) serving is fundamentally limited by the sequential nature of autoregressive decoding. Speculative Decoding (SD) mitigates this by using a lightweight draft...
The paper introduces the Expected Decoding Rounds (EDR) objective, a theoretical framework that models speculative decoding as a Markov reward process and directly optimizes the number of decoding rounds. EDR weights local rejection costs by state occupancies, eliminating auxiliary hyperparameters and enabling unbiased stochastic optimization via temporal‑difference gradients. Applying EDR to fine‑tune state‑of‑the‑art draft models (DSpark and DFly) improves mean accepted length and outperforms prior training objectives across nine benchmarks in math reasoning, code generation, and chat.
By Yunxiao Zhao, Changxiao Cai
Draft-OPD introduces an on‑policy distillation method for speculative draft models, addressing the mismatch between supervised fine‑tuning and inference by letting the target model supervise the drafter on draft‑induced states. The approach uses target‑assisted rollouts for stable continuations and replays drafting from error positions exposed during verification, enabling the drafter to learn from both accepted and rejected proposals. Experiments demonstrate that Draft‑OPD achieves more than five‑fold lossless acceleration across diverse tasks, outperforming prior draft models such as EAGLE‑3 and DFlash by 23 % and 13 % respectively.
By Haodi Lei, Yafu Li, Haoran Zhang, Shunkai Zhang, Qianjia Cheng, Xiaoye Qu, Ganqu Cui, Bowen Zhou, Ning Ding, Yun Luo, Yu Cheng
arXiv:2602. 05774v4 Announce Type: replace-cross Abstract: Speculative decoding accelerates inference for (M)LLMs, yet a training-decoding discrepancy persists: while existing methods optimize single greedy trajectories, decoding involves verifying and ranking multiple sampled draft paths.
By Xiandong Zou, Jianshu Li, Jing Huang, Pan Zhou
The paper introduces Parallel Reconsideration, a technique that lets a drafter in speculative decoding re-evaluate uncertain token guesses with additional parallel passes, adding minimal latency. This method, combined with a new training recipe and drafting algorithm, enhances the drafter’s acceptance rate and overall speedup. Experiments across models and benchmarks show average improvements of 5.3%–10.2% in acceptance length and 3.5%–9.2% in speedup.
By Hoang Anh Duy Le, Huibing Dong, Alireza Khadem, Ziteng Sun, Xiang Song, Yizheng Huang, Soroush Radpour, Anshumali Shrivastava, Ye He
The paper introduces Rank-Aware Speculative Sampling (RASS), a new verification rule for speculative draft trees in diffusion models. RASS orders draft candidates by proposal‑target mean displacement, samples a rank with weights optimized to minimize total variation, and then maximally couples the selected candidate with the target distribution, ensuring exact sampling. Experiments on Gaussian mixtures, FFHQ, CIFAR‑10, and Stable Diffusion 3.5 show that RASS reduces target‑model evaluation counts by up to 20% compared to Diffusion Greedy Rejection Sampling (D‑GRS).
By Marcello Bullo, Yanxiao Liu, \"Oyk\"u S{\i}la G\"uner, Arpan Mukherjee, Deniz G\"und\"uz
arXiv:2607.28077v2 Announce Type: replace
Abstract: Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models, but prompt groups with identica...
By Shuang Liang, Haoyang Zhou, Yifan Gong, Guowei Wang, Xiting Wang
ResiSpec is a framework that improves speculative decoding for large language models by reshaping the residual distribution during verification. It addresses the problem of residual drift, where rejected candidates cause the target distribution to diverge from the draft model’s predictions, rendering later candidates ineffective. By aligning the verification process with the draft model’s high‑confidence regions, ResiSpec prevents candidate obsolescence and achieves up to 1.92× speedup over existing multi‑candidate methods.
By Zhi-Kai Chen, Jun-Jie Tao, Wei-Xiang Mao, De-Chuan Zhan, Han-Jia Ye
The paper investigates how target model probability distributions can be reused during speculative decoding to improve token selection. It shows that exact reuse requires a distribution computed with the full next-round context, and that retaining this distribution can force the rejection of accepted tokens. Experiments across six target models demonstrate that reranking first-token branches with a separately computed distribution at the current context boosts expected acceptance, especially with small tree budgets, and that drawing the first token from this distribution before tree allocation further improves acceptance and decoding speed.
By Jungseob Lee, Giyoon Jang, Chailin Nam, Jiwoo Kum, Sihyeon Lee, Kyojun Choo, Jihun Song, Eunseo Song, Chanjun Park
Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models, but prompt groups with identical rollout rewards consume generation budget without effective learning signals. Pre-rollout prompt selection can reduce this waste by screening prompts before rollout generation.
arXiv:2603. 12617v2 Announce Type: replace-cross Abstract: Speculative decoding has emerged as a widely adopted paradigm for accelerating large language model inference, where a lightweight draft model rapidly generates candidate tokens that are then verified in parallel by a larger target model.
By Yu-Yang Qian, Hao-Cong Wu, Yichao Fu, Hao Zhang, Peng Zhao
arXiv:2608. 08721v1 Announce Type: cross Abstract: Speculative decoding accelerates large language model inference by drafting multiple tokens for parallel verification, with efficiency critically determined by the speculative length selected at each decoding round.
By Zexun Lin, Yuan Feng, Junlin Lv, Kevin S. Zhou, Xike Xie