English
Related papers

Related papers: Making Every Verified Token Count: Adaptive Verifi…

200 papers

Mixture-of-experts (MoE) is becoming popular due to its success in improving the model quality, especially in Transformers. By routing tokens with a sparse gate to a few experts (i.e., a small pieces of the full model), MoE can easily…

Machine Learning · Computer Science 2022-10-11 Xiaonan Nie , Xupeng Miao , Shijie Cao , Lingxiao Ma , Qibin Liu , Jilong Xue , Youshan Miao , Yi Liu , Zhi Yang , Bin Cui

Speculative decoding has emerged as a promising approach to accelerate autoregressive inference in large language models (LLMs). Self-draft methods, which leverage the base LLM itself for speculation, avoid the overhead of auxiliary draft…

Computation and Language · Computer Science 2026-04-15 Zhuofan Wen , Yang Feng

The recent advancements in large language models (LLMs) have been extraordinary, yet the escalating inference costs associated with them present challenges in real-world applications. To address these challenges, we propose a novel approach…

Computation and Language · Computer Science 2024-06-07 Jiahao Liu , Qifan Wang , Jingang Wang , Xunliang Cai

Speculative decoding accelerates generation by drafting candidates and verifying them in parallel, yet state-of-the-art drafters (e.g., EAGLE) still require N sequential passes to propose N tokens. We present FastEagle, a non-autoregressive…

Machine Learning · Computer Science 2025-09-26 Haiduo Huang , Jiangcheng Song , Wenzhe Zhao , Pengju Ren

The Mixture-of-Experts (MoE) architecture is showing promising results in improving parameter sharing in multi-task learning (MTL) and in scaling high-capacity neural networks. State-of-the-art MoE models use a trainable sparse gate to…

Speculative decoding accelerates inference in large language models (LLMs) by generating draft tokens for target model verification. Current approaches for obtaining draft tokens rely on lightweight draft models or additional model…

Computation and Language · Computer Science 2025-03-06 Guofeng Quan , Wenfeng Feng , Chuzhan Hao , Guochao Jiang , Yuewei Zhang , Hao Wang

The Mixture-of-Experts (MoE) architecture has emerged as a promising approach to mitigate the rising computational costs of large language models (LLMs) by selectively activating parameters. However, its high memory requirements and…

Artificial Intelligence · Computer Science 2026-04-14 Jehyeon Bang , Eunyeong Cho , Ranggi Hwang , Jinha Chung , Minsoo Rhu

Speculative decoding accelerates LLMs by using a lightweight draft model to generate tokens autoregressively before verifying them in parallel with a larger target model. However, determining the optimal number of tokens to draft remains a…

Machine Learning · Computer Science 2025-11-05 Aditya Sridhar , Nish Sinnadurai , Sean Lie , Vithursan Thangarasa

Large Language Models (LLMs) face significant inference latency challenges stemming from their autoregressive design and large size. To address this, speculative decoding emerges as a solution, enabling the simultaneous generation and…

Computation and Language · Computer Science 2026-02-27 Yinrong Hong , Zhiquan Tan , Kai Hu

Speculative decoding reduces the inference latency of a target large language model via utilizing a smaller and faster draft model. Its performance depends on a hyperparameter K -- the candidate length, i.e., the number of candidate tokens…

Computation and Language · Computer Science 2025-07-14 Kaixuan Huang , Xudong Guo , Mengdi Wang

Speculative decoding can substantially accelerate LLM inference, but realizing its benefits in practice is challenging due to evolving workloads and system-level constraints. We present TIDE (Temporal Incremental Draft Engine), a…

Machine Learning · Computer Science 2026-02-06 Jiyoung Park , Hankyu Jang , Changseok Song , Wookeun Jung

Current language models decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a…

Computation and Language · Computer Science 2024-06-04 Wenhong Zhu , Hongkun Hao , Zhiwei He , Yiming Ai , Rui Wang

Speculative decoding has emerged as a widely adopted method to accelerate large language model inference without sacrificing the quality of the model outputs. While this technique has facilitated notable speed improvements by enabling…

Computation and Language · Computer Science 2025-02-12 Jacob K Christopher , Brian R Bartoldson , Tal Ben-Nun , Michael Cardei , Bhavya Kailkhura , Ferdinando Fioretto

Autoregressive decoding is bottlenecked by its sequential nature. Speculative decoding has become a standard way to accelerate inference by using a fast draft model to predict upcoming tokens from a slower target model, and then verifying…

Machine Learning · Computer Science 2026-05-06 Tanishq Kumar , Tri Dao , Avner May

Speculative decoding is a technique to leverage hardware concurrency in order to enable multiple steps of token generation in a single forward pass, thus improving the efficiency of large-scale autoregressive (AR) Transformer models.…

Machine Learning · Computer Science 2025-10-29 Yangchao Wu , Zongyue Qin , Alex Wong , Stefano Soatto

Speculative decoding accelerates LLM inference but suffers from performance degradation when target models are fine-tuned for specific domains. A naive solution is to retrain draft models for every target model, which is costly and…

Machine Learning · Computer Science 2026-03-11 Luxi Lin , Zhihang Lin , Zhanpeng Zeng , Yuhao Chen , Qingyu Zhang , Jixiang Luo , Xuelong Li , Rongrong Ji

Mixture-of-Experts (MoE) architectures enable efficient scaling of large language models by activating only a subset of parameters per input. However, existing MoE models suffer from two critical limitations: (1) inefficient token-to-expert…

Computation and Language · Computer Science 2025-10-10 Jing Li , Zhijie Sun , Dachao Lin , Xuan He , Binfan Zheng , Yi Lin , Rongqian Zhao , Xin Chen

Recent works have revealed the great potential of speculative decoding in accelerating the autoregressive generation process of large language models. The success of these methods relies on the alignment between draft candidates and the…

Computation and Language · Computer Science 2025-09-15 Jikai Wang , Zhenxu Tian , Juntao Li , Qingrong Xia , Xinyu Duan , Zhefeng Wang , Baoxing Huai , Min Zhang

The Mixture-of-Experts (MoE) architecture has been widely adopted in large language models (LLMs) to reduce computation cost through model sparsity. Employing speculative decoding (SD) can further accelerate MoE inference by drafting…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-07 Liangkun Chen , Zijian Wen , Tian Wu , Xiaoxi Zhang , Chuan Wu

Test-time scaling improves LLM performance by generating multiple candidate solutions, yet token-level sampling requires temperature tuning that trades off diversity against stability. Fine-grained MoE, featuring hundreds of well-trained…

Machine Learning · Computer Science 2026-05-04 Yuanteng Chen , Peisong Wang , Nanxin Zeng , Yuantian Shao , Shuang Qiu , Gang Li , Jing Liu , Jian Cheng