English
Related papers

Related papers: MoE-Spec: Expert Budgeting for Efficient Speculati…

200 papers

Large Language Models (LLMs) have become essential in advancing natural language processing (NLP) tasks, but their sequential token generation limits inference speed. Multi-Draft Speculative Decoding (MDSD) offers a promising solution by…

Computation and Language · Computer Science 2024-11-11 Ryan Sun , Tianyi Zhou , Xun Chen , Lichao Sun

Quantization is widely adopted to accelerate inference and reduce memory consumption in large language models (LLMs). While activation-weight joint quantization enables efficient low-precision decoding, it suffers from substantial…

Machine Learning · Computer Science 2025-10-03 Juntao Zhao , Wenhao Lu , Sheng Wang , Lingpeng Kong , Chuan Wu

Mixture-of-Experts (MoE) architectures are widely used in modern large language models and multimodal models. However, inference efficiency is often limited by highly dynamic and skewed expert workloads across different modalities. During…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-12 Yingping Wang , Yi Wu , Xiangyu Wu , Junwei Cui , Weilin Cai , Zhijiang Guo , Jiayi Huang

Vision-language models (VLMs) achieve strong performance on multimodal tasks but suffer from high inference latency due to large model sizes and long multimodal contexts. Speculative decoding has recently emerged as an effective…

Computer Vision and Pattern Recognition · Computer Science 2026-03-17 Hui Shen , Xin Wang , Ping Zhang , Yunta Hsieh , Qi Han , Zhongwei Wan , Ziheng Zhang , Jingxuan Zhang , Jing Xiong , Ziyuan Liu , Yifan Zhang , Hangrui Cao , Chenyang Zhao , Mi Zhang

Mixture of Experts (MoE) LLMs have recently gained attention for their ability to enhance performance by selectively engaging specialized subnetworks or "experts" for each input. However, deploying MoEs on memory-constrained devices remains…

Speculative decoding is an effective method for lossless acceleration of large language models during inference. It uses a fast model to draft a block of tokens which are then verified in parallel by the target model, and provides a…

Machine Learning · Computer Science 2025-04-14 Ziteng Sun , Uri Mendlovic , Yaniv Leviathan , Asaf Aharoni , Jae Hun Ro , Ahmad Beirami , Ananda Theertha Suresh

The promising applications of large language models are often limited by the constrained GPU memory capacity available on edge devices. Mixture-of-Experts (MoE) models help address this issue by activating only a subset of the model's…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-09-03 Xiaoniu Song , Zihang Zhong , Rong Chen , Haibo Chen

The sparsely activated mixture-of-experts (MoE) transformer has become a common architecture for large language models (LLMs) due to its sparsity, which requires fewer computational demands while easily scaling the model size. In MoE…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-14 Wenxiang Lin , Xinglin Pan , Lin Zhang , Shaohuai Shi , Xuan Wang , Xiaowen Chu

Mixture-of-Experts (MoE) has emerged as a prominent architecture for scaling model size while maintaining computational efficiency. In MoE, each token in the input sequence activates a different subset of experts determined by a routing…

Computation and Language · Computer Science 2024-11-05 Chufan Shi , Cheng Yang , Xinyu Zhu , Jiahao Wang , Taiqiang Wu , Siheng Li , Deng Cai , Yujiu Yang , Yu Meng

Estimating accurate and well-calibrated predictive uncertainty is important for enhancing the reliability of computer vision models, especially in safety-critical applications like traffic scene perception. While ensemble methods are…

Computer Vision and Pattern Recognition · Computer Science 2025-09-08 Svetlana Pavlitska , Beyza Keskin , Alwin Faßbender , Christian Hubschneider , J. Marius Zöllner

Mixture-of-Experts (MoE) models enable sparse expert activation, meaning that only a subset of the model's parameters is used during each inference. However, to translate this sparsity into practical performance, an expert caching mechanism…

Machine Learning · Computer Science 2026-02-05 Duc Hoang , Ajay Jaiswal , Mohammad Samragh , Minsik Cho

Speculative decoding is a standard method for accelerating the inference speed of large language models. However, scaling it for production environments poses several engineering challenges, including efficiently implementing different…

The advancement of deep learning has led to the emergence of Mixture-of-Experts (MoEs) models, known for their dynamic allocation of computational resources based on input. Despite their promise, MoEs face challenges, particularly in terms…

Computation and Language · Computer Science 2024-04-09 Alexandre Muzio , Alex Sun , Churan He

Cloud-based Large Language Model (LLM) services often face challenges in achieving low inference latency and meeting Service Level Objectives (SLOs) under dynamic request patterns. Speculative decoding, which exploits lightweight models for…

Computation and Language · Computer Science 2026-01-13 Kaiyu Huang , Hao Wu , Zhubo Shi , Han Zou , Minchen Yu , Qingjiang Shi

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,…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-15 Yuhao Shen , Junyi Shen , Quan Kong , Tianyu Liu , Yao Lu , Cong Wang

The Mixtures-of-Experts (MoE) model is a widespread distributed and integrated learning method for large language models (LLM), which is favored due to its ability to sparsify and expand models efficiently. However, the performance of MoE…

Machine Learning · Computer Science 2024-05-24 Jing Li , Zhijie Sun , Xuan He , Li Zeng , Yi Lin , Entong Li , Binfan Zheng , Rongqian Zhao , Xin Chen

Inference with Multimodal Large Language Models (MLLMs) is slow due to their large-language-model backbone which suffers from memory bandwidth bottleneck and generates tokens auto-regressively. In this paper, we explore the application of…

Computation and Language · Computer Science 2024-04-16 Mukul Gagrani , Raghavv Goel , Wonseok Jeon , Junyoung Park , Mingu Lee , Christopher Lott

Transformer-based autoregressive sampling has been the major bottleneck for slowing down large language model inferences. One effective way to accelerate inference is \emph{Speculative Decoding}, which employs a small model to sample a…

Machine Learning · Computer Science 2024-11-05 Ming Yin , Minshuo Chen , Kaixuan Huang , Mengdi Wang

Large language models and large multimodal models (LLMs and LMMs) deliver strong generative performance but suffer from slow decoding, a problem that becomes more severe when handling visual inputs, whose sequences typically contain many…

Computer Vision and Pattern Recognition · Computer Science 2026-02-04 Zihua Wang , Ruibo Li , Haozhe Du , Joey Tianyi Zhou , Yu Zhang , Xu Yang

Mixture-of-Experts (MoE) Large Language Models (LLMs) face a trilemma of load imbalance, parameter redundancy, and communication overhead. We introduce a unified framework based on dynamic expert clustering and structured compression to…

Computation and Language · Computer Science 2026-02-06 Peijun Zhu , Ning Yang , Baoliang Tian , Jiayu Wei , Weihao Zhang , Haijun Zhang , Pin Lv