English
Related papers

Related papers: Taming Sparsely Activated Transformer with Stochas…

200 papers

Sparse Mixture-of-Experts (MoE) allows scaling of language and vision models efficiently by activating only a small subset of experts per input. While this reduces computation, the large number of parameters still incurs substantial memory…

Sparsely-gated mixture-of-experts (MoE) has been widely adopted to scale deep learning models to trillion-plus parameters with fixed computational cost. The algorithmic performance of MoE relies on its token routing mechanism that forwards…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-06-06 Changho Hwang , Wei Cui , Yifan Xiong , Ziyue Yang , Ze Liu , Han Hu , Zilong Wang , Rafael Salas , Jithin Jose , Prabhat Ram , Joe Chau , Peng Cheng , Fan Yang , Mao Yang , Yongqiang Xiong

We present Marco-MoE, a suite of fully open multilingual sparse Mixture-of-Experts (MoE) models. Marco-MoE features a highly sparse design in which only around 5\% of the total parameters are activated per input token. This extreme…

Computation and Language · Computer Science 2026-04-29 Fan Jiang , Yu Zhao , Chenyang Lyu , Tianqi Shi , Yichao Du , Feihu Jiang , Longyue Wang , Weihua Luo

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

Sparsely activated Mixture-of-Experts (SMoE) has shown promise to scale up the learning capacity of neural networks, however, they have issues like (a) High Memory Usage, due to duplication of the network layers into multiple copies as…

Machine Learning · Computer Science 2024-03-15 Pingzhi Li , Zhenyu Zhang , Prateek Yadav , Yi-Lin Sung , Yu Cheng , Mohit Bansal , Tianlong Chen

Scaling up the number of parameters of language models has proven to be an effective approach to improve performance. For dense models, increasing model size proportionally increases the model's computation footprint. In this work, we seek…

Computation and Language · Computer Science 2023-11-21 Cicero Nogueira dos Santos , James Lee-Thorp , Isaac Noble , Chung-Ching Chang , David Uthus

Time series forecasting models are increasingly scaled through large Transformer backbones, yet most existing approaches process all series through a shared dense computation path despite substantial heterogeneity in temporal structure.…

Machine Learning · Computer Science 2026-05-26 Rui Wang , Renhao Xue , Ray Razi , Huan Song , Hannah R. Marlowe

Mixture-of-Experts (MoE) architectures have become the key to scaling modern LLMs, yet little is understood about how their sparse routing dynamics respond to multilingual data. In this work, we analyze expert routing patterns using…

Computation and Language · Computer Science 2026-02-19 Lucas Bandarkar , Chenyuan Yang , Mohsen Fayyaz , Junlin Hu , Nanyun Peng

Sparse mixture of experts (SMoE) have emerged as an effective approach for scaling large language models while keeping a constant computational cost. Regardless of several notable successes of SMoE, effective training such architecture…

Computation and Language · Computer Science 2024-06-25 Giang Do , Hung Le , Truyen Tran

Large language models, such as OpenAI's ChatGPT, have demonstrated exceptional language understanding capabilities in various NLP tasks. Sparsely activated mixture-of-experts (MoE) has emerged as a promising solution for scaling models…

Computation and Language · Computer Science 2023-10-12 Jiamin Li , Qiang Su , Yitao Yang , Yimin Jiang , Cong Wang , Hong Xu

Sparse Upcycling provides an efficient way to initialize a Mixture-of-Experts (MoE) model from pretrained dense weights instead of training from scratch. However, since all experts start from identical weights and the router is randomly…

Computer Vision and Pattern Recognition · Computer Science 2026-04-20 Sanghyeok Chu , Pyunghwan Ahn , Gwangmo Song , SeungHwan Kim , Honglak Lee , Bohyung Han

Sparse Mixture-of-Experts (MoE) models scale parameters while fixing active computation per token, but the specialization of individual experts remains opaque. In a companion paper we showed that routing topology is quality-neutral: five…

Artificial Intelligence · Computer Science 2026-04-17 Ivan Ternovtsii , Yurii Bilak

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

Standard LoRA fine-tuning of Mixture-of-Experts (MoE) models applies adapters to every expert, yet our profiling shows that per-layer expert routing is highly skewed: a small subset of experts handles most tokens in each layer, while many…

Machine Learning · Computer Science 2026-03-26 Andrea Manzoni

Recently, mixture of experts (MoE) has become a popular paradigm for achieving the trade-off between modal capacity and efficiency of multi-modal large language models (MLLMs). Different from previous efforts, we are dedicated to exploring…

Multimedia · Computer Science 2025-02-13 Qiong Wu , Zhaoxi Ke , Yiyi Zhou , Xiaoshuai Sun , Rongrong Ji

Mixture-of-Experts (MoE) has emerged as a powerful paradigm for scaling model capacity while preserving computational efficiency. Despite its notable success in large language models (LLMs), existing attempts to apply MoE to Diffusion…

Computer Vision and Pattern Recognition · Computer Science 2026-03-03 Yujie Wei , Shiwei Zhang , Hangjie Yuan , Yujin Han , Zhekai Chen , Jiayu Wang , Difan Zou , Xihui Liu , Yingya Zhang , Yu Liu , Hongming Shan

Mixture-of-Experts (MoE) architectures enable conditional computation by routing inputs to multiple expert subnetworks and are often motivated as a mechanism for scaling large language models. In this project, we instead study MoE behavior…

Machine Learning · Computer Science 2026-01-22 Adam Rokah , Daniel Veress , Caleb Caulk , Sourav Sharan

Most recent state-of-the-art (SOTA) large language models (LLMs) use Mixture-of-Experts (MoE) architectures to scale model capacity without proportional per-token compute, enabling higher-quality outputs at manageable serving costs.…

The Mixture of Experts (MoE) is an advanced model architecture in the industry that combines multiple specialized expert models from various domains into a single supermodel. This approach enables the model to scale without significantly…

Machine Learning · Computer Science 2024-11-04 Jingming Guo , Yan Liu , Yu Meng , Zhiwei Tao , Banglan Liu , Gang Chen , Xiang Li

Mixture-of-Experts (MoE) has emerged as a promising paradigm for efficiently scaling large language models without a proportional increase in computational cost. However, the standard training strategy of Top-K router prevents MoE models…

Computation and Language · Computer Science 2025-10-01 Yaoxiang Wang , Qingguo Hu , Yucheng Ding , Ruizhe Wang , Yeyun Gong , Jian Jiao , Yelong Shen , Peng Cheng , Jinsong Su