English
Related papers

Related papers: Route Experts by Sequence, not by Token

200 papers

Mixture of Experts (MoE) has become a key architectural paradigm for efficiently scaling Large Language Models (LLMs) by selectively activating a subset of parameters for each input token. However, standard MoE architectures face…

Machine Learning · Computer Science 2025-05-27 Zehua Liu , Han Wu , Ruifeng She , Xiaojin Fu , Xiongwei Han , Tao Zhong , Mingxuan Yuan

Mixture-of-Experts (MoE) models have demonstrated exceptional performance in large-scale language models. Existing routers typically rely on non-differentiable Top-$k$+Softmax, limiting their performance and scalability. We argue that two…

We introduce MixtureKit, a modular open-source framework for constructing, training, and analyzing Mixture-of-Experts (MoE) models from arbitrary pre-trained or fine-tuned models. MixtureKit currently supports three complementary methods:…

Machine Learning · Computer Science 2025-12-16 Ahmad Chamma , Omar El Herraoui , Guokan Shang

Sparse Mixture-of-Experts (MoE) architectures effectively scale model capacity by activating only a subset of experts for each input token. However, the standard Top-k routing strategy imposes a uniform sparsity pattern that ignores the…

Artificial Intelligence · Computer Science 2025-12-17 Can Jin , Hongwu Peng , Mingcan Xiang , Qixin Zhang , Xiangchi Yuan , Amit Hasan , Ohiremen Dibua , Yifan Gong , Yan Kang , Dimitris N. Metaxas

Mixture of Experts (MoE) architectures increase large language model scalability, yet their performance depends on the router module that moves tokens to specialized experts. Bad routing can load imbalance and reduced accuracy. This project…

Machine Learning · Computer Science 2025-06-23 Daniel Fidel Harvey , George Weale , Berk Yilmaz

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

Mixture-of-Experts (MoE) models have emerged as a dominant paradigm for efficient LLM scaling, yet adapting them to non-English downstream tasks remains challenging. Existing fine-tuning approaches treat MoE models as monolithic learners,…

Computation and Language · Computer Science 2026-05-28 Guanzhi Deng , Kuan Wu , Haibo Wang , Shing Yin Wong , Sichun Luo , Linqi Song

Mixture-of-Experts (MoE) architectures enable efficient model scaling, yet expert routing behavior across underrepresented languages remains poorly understood. We analyze routing dynamics in two architecturally distinct MoE models -- a pure…

Computation and Language · Computer Science 2026-05-19 Ori Bar Joseph , Smadar Arvatz , Noam Kayzer , Dan Revital , Sarel Weinberger

Sparse Mixture-of-Experts (MoE) architectures are increasingly popular for frontier large language models (LLM) but they introduce training challenges due to routing complexity. Fully leveraging parameters of an MoE model requires all…

Sparse Mixture-of-Experts (SMoE) has gained prominence as a scalable and computationally efficient architecture, enabling significant growth in model capacity without incurring additional inference costs. However, existing SMoE models often…

Machine Learning · Computer Science 2025-11-13 Duc Anh Nguyen , Huu Binh Ta , Nhuan Le Duc , Tan M. Nguyen , Toan Tran

Mixture-of-Experts Large Language Models (MoE-LLMs) achieve strong performance but incur substantial memory overhead due to massive expert parameters. Mixed-precision quantization mitigates this cost by allocating expert-wise bit-widths…

Machine Learning · Computer Science 2026-05-25 Jianing Deng , Song Wang , Dongwei Wang , Zijie Liu , Tianlong Chen , Huanrui Yang , Jingtong Hu

Open-source Large Language Models (LLMs) increasingly specialize by domain (e.g., math, code, general reasoning), motivating systems that leverage complementary strengths across models. Prior multi-LLM approaches either (i) route a query to…

Machine Learning · Computer Science 2025-09-26 Jacob Fein-Ashley , Dhruv Parikh , Rajgopal Kannan , Viktor Prasanna

This study investigates the behavior of model-integrated routers in Mixture of Experts (MoE) models, focusing on how tokens are routed based on their linguistic features, specifically Part-of-Speech (POS) tags. The goal is to explore across…

Computation and Language · Computer Science 2024-12-24 Elie Antoine , Frédéric Béchet , Philippe Langlais

Mixture of Experts (MoE) models have emerged as a promising paradigm for scaling language models efficiently by activating only a subset of parameters for each input token. In this report, we present dots.llm1, a large-scale MoE model that…

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

Model editing aims to efficiently alter the behavior of Large Language Models (LLMs) within a desired scope, while ensuring no adverse impact on other inputs. Recent years have witnessed various model editing methods been proposed. However,…

Computation and Language · Computer Science 2024-06-04 Renzhi Wang , Piji Li

Training large language models (LLMs) for different inference constraints is computationally expensive, limiting control over efficiency-accuracy trade-offs. Moreover, once trained, these models typically process tokens uniformly,…

Computation and Language · Computer Science 2025-02-19 Kumari Nishu , Sachin Mehta , Samira Abnar , Mehrdad Farajtabar , Maxwell Horton , Mahyar Najibi , Moin Nabi , Minsik Cho , Devang Naik

Mixture-of-Experts (MoE) enhances model performance while maintaining computational efficiency, making it well-suited for large-scale applications. Conventional mixture-of-experts (MoE) architectures suffer from suboptimal coordination…

Machine Learning · Computer Science 2025-09-24 Yujiao Yang , Jing Lian , Linhui Li

Mixture-of-Experts (MoE) has been demonstrated as an efficient method to scale up models. By dynamically and sparsely selecting activated experts, MoE can effectively reduce computational costs. Despite the success, we observe that many…

Machine Learning · Computer Science 2024-06-19 Haoze Wu , Zihan Qiu , Zili Wang , Hang Zhao , Jie Fu

The mixture-of-experts (MoE) architecture, which replaces dense networks with sparse ones, has attracted significant attention in large vision-language models (LVLMs) for achieving comparable performance while activating far fewer…

Computer Vision and Pattern Recognition · Computer Science 2026-04-03 Chaoxiang Cai , Longrong Yang , Minghe Weng , Xuewei Li , Zequn Qin , Xi Li