English
Related papers

Related papers: $\mu$-MoE: Test-Time Pruning as Micro-Grained Mixt…

200 papers

Mixture-of-Experts (MoE) models scale capacity efficiently, but their massive parameter footprint creates a deployment-time memory bottleneck. We organize retraining-free MoE compression into three paradigms - Expert Pruning, Expert…

Machine Learning · Computer Science 2026-03-04 Sieun Hyeon , Jaeyoung Do

In this work, we first explore whether the parameters activated by the MoE layer remain highly sparse at inference. We perform a sparsification study on several representative MoE models. For each expert, we rank parameters by the magnitude…

Computation and Language · Computer Science 2025-10-08 Runxi Cheng , Yuchen Guan , Yucheng Ding , Qingguo Hu , Yongxian Wei , Chun Yuan , Yelong Shen , Weizhu Chen , Yeyun Gong

Mixture-of-Experts (MoE) models typically fix the number of activated experts $k$ at both training and inference. However, real-world deployments often face heterogeneous hardware, fluctuating workloads, and diverse quality-latency…

Computation and Language · Computer Science 2026-05-12 Naibin Gu , Zhenyu Zhang , Yuchen Feng , Yilong Chen , Peng Fu , Zheng Lin , Shuohuan Wang , Yu Sun , Hua Wu , Weiping Wang , Haifeng Wang

Sparsely-activated Mixture-of-Experts (SMoE) models offer efficient pre-training and low latency but their large parameter counts create significant memory overhead, motivating research into expert compression. Contrary to recent findings…

Machine Learning · Computer Science 2026-05-14 Mike Lasby , Ivan Lazarevich , Nish Sinnadurai , Sean Lie , Yani Ioannou , Vithursan Thangarasa

Mixture-of-Experts (MoE) models offer dynamic computation, but are typically deployed as static full-capacity models, missing opportunities for deployment-specific specialization. We introduce PreMoE, a training-free framework that…

Machine Learning · Computer Science 2026-04-27 Zehua Pei , Ying Zhang , Hui-Ling Zhen , Tao Yuan , Xianzhi Yu , Zhenhua Dong , Sinno Jialin Pan , Mingxuan Yuan , Bei Yu

Mixture-of-Experts large language models (MoE-LLMs) marks a significant step forward of language models, however, they encounter two critical challenges in practice: 1) expert parameters lead to considerable memory consumption and loading…

Machine Learning · Computer Science 2025-02-25 Wei Huang , Yue Liao , Jianhui Liu , Ruifei He , Haoru Tan , Shiming Zhang , Hongsheng Li , Si Liu , Xiaojuan Qi

Mixture-of-Experts (MoE) is a promising way to scale up the learning capacity of large language models. It increases the number of parameters while keeping FLOPs nearly constant during inference through sparse activation. Yet, it still…

Machine Learning · Computer Science 2025-02-26 Pingzhi Li , Xiaolong Jin , Zhen Tan , Yu Cheng , Tianlong Chen

Sparse Mixture of Expert (SMoE) models have emerged as a scalable alternative to dense models in language modeling. These models use conditionally activated feedforward subnetworks in transformer blocks, allowing for a separation between…

Machine Learning · Computer Science 2024-09-04 Soumajyoti Sarkar , Leonard Lausen , Volkan Cevher , Sheng Zha , Thomas Brox , George Karypis

The emergence of Mixture of Experts (MoE) LLMs has significantly advanced the development of language models. Compared to traditional LLMs, MoE LLMs outperform traditional LLMs by achieving higher performance with considerably fewer…

Machine Learning · Computer Science 2024-11-05 Cheng Yang , Yang Sui , Jinqi Xiao , Lingyi Huang , Yu Gong , Yuanlin Duan , Wenqi Jia , Miao Yin , Yu Cheng , Bo Yuan

Mixture-of-Experts (MoE) architectures achieve parameter efficiency through conditional computation, yet contemporary designs suffer from two fundamental limitations: structural parameter isolation that causes catastrophic forgetting, and…

Machine Learning · Computer Science 2026-01-21 Yuxing Gan , Ziyu Lei

The Mixture of Experts (MoE) architecture is an important method for scaling Large Language Models (LLMs). It increases model capacity while keeping computation cost low. However, the ultra-large MoE models still have hundreds of billions…

Artificial Intelligence · Computer Science 2025-10-01 Yixiao Chen , Yanyue Xie , Ruining Yang , Wei Jiang , Wei Wang , Yong He , Yue Chen , Pu Zhao , Yanzhi Wang

Sparsely activated Mixture-of-Experts (SMoE) has shown promise in scaling up the learning capacity of neural networks. However, vanilla SMoEs have issues such as expert redundancy and heavy memory requirements, making them inefficient and…

Machine Learning · Computer Science 2025-04-11 Ajay Jaiswal , Jianyu Wang , Yixiao Li , Pingzhi Li , Tianlong Chen , Zhangyang Wang , Chong Wang , Ruoming Pang , Xianzhi Du

Mixture-of-Experts (MoE) models scale large language models efficiently by sparsely activating experts, but once an expert is selected, it is executed fully. Hence, the trade-off between accuracy and computation in an MoE model typically…

Machine Learning · Computer Science 2026-02-09 Nurbek Tastan , Stefanos Laskaridis , Karthik Nandakumar , Samuel Horvath

A pivotal advancement in the progress of large language models (LLMs) is the emergence of the Mixture-of-Experts (MoE) LLMs. Compared to traditional LLMs, MoE LLMs can achieve higher performance with fewer parameters, but it is still hard…

Computation and Language · Computer Science 2024-05-31 Xudong Lu , Qi Liu , Yuhui Xu , Aojun Zhou , Siyuan Huang , Bo Zhang , Junchi Yan , Hongsheng Li

In this paper, we introduce a novel dynamic expert selection framework for Mixture of Experts (MoE) models, aiming to enhance computational efficiency and model performance by adjusting the number of activated experts based on input…

Machine Learning · Computer Science 2024-03-13 Quzhe Huang , Zhenwei An , Nan Zhuang , Mingxu Tao , Chen Zhang , Yang Jin , Kun Xu , Kun Xu , Liwei Chen , Songfang Huang , Yansong Feng

The Mixture of Experts (MoE) is a widely known neural architecture where an ensemble of specialized sub-models optimizes overall performance with a constant computational cost. However, conventional MoEs pose challenges at scale due to the…

Computation and Language · Computer Science 2023-09-12 Ted Zadouri , Ahmet Üstün , Arash Ahmadian , Beyza Ermiş , Acyr Locatelli , Sara Hooker

Mixture-of-Experts (MoE) models have shown strong potential in scaling language models efficiently by activating only a small subset of experts per input. However, their widespread deployment remains limited due to the high memory overhead…

Machine Learning · Computer Science 2025-11-10 Yushu Zhao , Zheng Wang , Minjia Zhang

Large language models are typically deployed as monolithic systems, requiring the full model even when applications need only a narrow subset of capabilities, e.g., code, math, or domain-specific knowledge. Mixture-of-Experts (MoEs)…

Computation and Language · Computer Science 2026-05-12 Ryan Wang , Akshita Bhagia , Sewon Min

Large multimodal models have achieved remarkable progress in both understanding and generation. Recent efforts pursue unified multimodal models that integrate heterogeneous components to support both capabilities within a single framework.…

Computer Vision and Pattern Recognition · Computer Science 2025-12-03 Shwai He , Chaorui Deng , Ang Li , Shen Yan

Mixture-of-Experts (MoE) has demonstrated promising potential in scaling LLMs. However, it is hindered by two critical challenges: (1) substantial GPU memory consumption to load all experts; (2) low activated parameters cannot be…

Machine Learning · Computer Science 2025-08-05 Yuanteng Chen , Yuantian Shao , Peisong Wang , Jian Cheng