English
Related papers

Related papers: Efficient Quantization of Mixture-of-Experts with …

200 papers

The Mixture-of-Experts (MoE) architecture has become a predominant paradigm for scaling large language models (LLMs). Despite offering strong performance and computational efficiency, large MoE-based LLMs like DeepSeek-V3-0324 and…

Machine Learning · Computer Science 2025-08-08 Xiaodong Chen , Mingming Ha , Zhenzhong Lan , Jing Zhang , Jianguo Li

Sparse Mixture-of-Experts (MoE) models scale capacity by routing each token to a small subset of experts. However, their routers exhibit a fundamental trade-off: strong load balancing can suppress expert specialization, while aggressive…

Machine Learning · Computer Science 2026-05-12 Gleb Molodtsov , Alexander Miasnikov , Aleksandr Beznosikov

Mixture-of-Experts (MoE) has become a practical architecture for scaling LLM capacity while keeping per-token compute modest, but deploying MoE models on a single, memory-limited GPU remains difficult because expert weights dominate the HBM…

Performance · Computer Science 2026-02-09 Kexin Chu , Dawei Xiang , Zixu Shen , Yiwei Yang , Zecheng Liu , Wei Zhang

The emergence of large-scale Mixture of Experts (MoE) models represents a significant advancement in artificial intelligence, offering enhanced model capacity and computational efficiency through conditional computation. However, deploying…

Machine Learning · Computer Science 2025-01-23 Jiacheng Liu , Peng Tang , Wenfeng Wang , Yuhang Ren , Xiaofeng Hou , Pheng-Ann Heng , Minyi Guo , Chao Li

The Mixture of Experts (MoE) models are an emerging class of sparsely activated deep learning models that have sublinear compute costs with respect to their parameters. In contrast with dense models, the sparse architecture of MoE offers…

A critical approach for efficiently deploying Mixture-of-Experts (MoE) models with massive parameters is quantization. However, state-of-the-art MoE models suffer from non-negligible accuracy loss with extreme quantization, such as under 4…

Machine Learning · Computer Science 2025-04-08 Beichen Huang , Yueming Yuan , Zelei Shao , Minjia Zhang

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

We propose a novel adaptive Mixture-of-Experts (MoE) framework for time series forecasting that enhances expert specialization by incorporating expert-specific loss information directly into the training process. Notably, the overall…

Machine Learning · Statistics 2026-05-12 Btissame El Mahtout , Florian Ziel

The Mixture-of-Experts (MoE) architecture enables a significant increase in the total number of model parameters with minimal computational overhead. However, it is not clear what performance tradeoffs, if any, exist between MoEs and…

While transformers and their variant conformers show promising performance in speech recognition, the parameterized property leads to much memory cost during training and inference. Some works use cross-layer weight-sharing to reduce the…

Audio and Speech Processing · Electrical Eng. & Systems 2022-09-20 Ye Bai , Jie Li , Wenjing Han , Hao Ni , Kaituo Xu , Zhuo Zhang , Cheng Yi , Xiaorui Wang

Mixture-of-Experts (MoE) models have recently demonstrated exceptional performance across a diverse range of applications. The principle of sparse activation in MoE models facilitates an offloading strategy, wherein active experts are…

Computation and Language · Computer Science 2025-10-15 Yushu Zhao , Yubin Qin , Yang Wang , Xiaolong Yang , Huiming Han , Shaojun Wei , Yang Hu , Shouyi Yin

Neural network quantization is frequently used to optimize model size, latency and power consumption for on-device deployment of neural networks. In many cases, a target bit-width is set for an entire network, meaning every layer get…

Machine Learning · Computer Science 2023-02-13 Nilesh Prasad Pandey , Markus Nagel , Mart van Baalen , Yin Huang , Chirag Patel , Tijmen Blankevoort

Mixture-of-Expert (MoE) based large language models (LLMs), such as the recent Mixtral and DeepSeek-MoE, have shown great promise in scaling model size without suffering from the quadratic growth of training cost of dense transformers. Like…

Machine Learning · Computer Science 2024-04-04 Longfei Yun , Yonghao Zhuang , Yao Fu , Eric P Xing , Hao 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

Neurons in large language models often exhibit \emph{polysemanticity}, simultaneously encoding multiple unrelated concepts and obscuring interpretability. Instead of relying on post-hoc methods, we present \textbf{MoE-X}, a…

Sparse Mixture-of-Experts (SMoE) architectures have gained prominence for their ability to scale neural networks, particularly transformers, without a proportional increase in computational cost. Despite their success, their role in…

Sparse Mixtures of Experts (SMoE) scales model capacity without significant increases in training and inference costs, but exhibits the following two issues: (1) Low expert activation, where only a small subset of experts are activated for…

Computation and Language · Computer Science 2024-04-24 Xun Wu , Shaohan Huang , Wenhui Wang , Furu Wei

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

Recent advances in large language models highlighted the excessive quadratic cost of self-attention. Despite the significant research efforts, subquadratic attention methods still suffer from inferior performance in practice. We hypothesize…

Machine Learning · Computer Science 2025-05-02 Piotr Piękos , Róbert Csordás , Jürgen Schmidhuber

Mixture-of-Experts (MoE) models are a promising way to scale up model capacity without significantly increasing computational cost. A key component of MoEs is the router, which decides which subset of parameters (experts) process which…

Computer Vision and Pattern Recognition · Computer Science 2024-04-22 Tianlin Liu , Mathieu Blondel , Carlos Riquelme , Joan Puigcerver