中文
相关论文

相关论文: Theory on Mixture-of-Experts in Continual Learning

200 篇论文

In deep learning, mixture-of-experts (MoE) activates one or few experts (sub-networks) on a per-sample or per-token basis, resulting in significant computation reduction. The recently proposed \underline{p}atch-level routing in…

机器学习 · 计算机科学 2023-07-10 Mohammed Nowaz Rabbani Chowdhury , Shuai Zhang , Meng Wang , Sijia Liu , Pin-Yu Chen

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…

机器学习 · 计算机科学 2025-09-24 Yujiao Yang , Jing Lian , Linhui Li

Mixture-of-Experts (MoE) models offer immense capacity via sparsely gated expert subnetworks, yet adapting them to multiple domains without catastrophic forgetting remains an open challenge. Existing approaches either incur prohibitive…

机器学习 · 计算机科学 2025-09-23 Junzhuo Li , Bo Wang , Xiuze Zhou , Xuming Hu

Mixture-of-Experts (MoE) networks have been proposed as an efficient way to scale up model capacity and implement conditional computing. However, the study of MoE components mostly focused on the feedforward layer in Transformer…

计算与语言 · 计算机科学 2022-10-12 Xiaofeng Zhang , Yikang Shen , Zeyu Huang , Jie Zhou , Wenge Rong , Zhang Xiong

Optimizing various wireless user tasks poses a significant challenge for networking systems because of the expanding range of user requirements. Despite advancements in Deep Reinforcement Learning (DRL), the need for customized optimization…

网络与互联网体系结构 · 计算机科学 2024-02-16 Hongyang Du , Guangyuan Liu , Yijing Lin , Dusit Niyato , Jiawen Kang , Zehui Xiong , Dong In Kim

Rapid advancements in deep learning have led to many recent breakthroughs. While deep learning models achieve superior performance, often statistically better than humans, their adoption into safety-critical settings, such as healthcare or…

机器学习 · 计算机科学 2022-04-08 Marko Vasic , Andrija Petrovic , Kaiyuan Wang , Mladen Nikolic , Rishabh Singh , Sarfraz Khurshid

The Mixture of Experts (MoE) architecture has become a fundamental building block in state-of-the-art large language models (LLMs), improving domain-specific expertise in LLMs and scaling model capacity without proportionally increasing…

机器学习 · 计算机科学 2026-05-13 Ankit Jyothish , Ali Jannesari , Aishwarya Sarkar , Joseph Zuber

A useful strategy to deal with complex classification scenarios is the "divide and conquer" approach. The mixture of experts (MOE) technique makes use of this strategy by joinly training a set of classifiers, or experts, that are…

机器学习 · 计算机科学 2014-05-30 Billy Peralta

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…

机器学习 · 计算机科学 2024-05-24 Jing Li , Zhijie Sun , Xuan He , Li Zeng , Yi Lin , Entong Li , Binfan Zheng , Rongqian Zhao , Xin Chen

Mixture-of-experts (MoE) model incorporates the power of multiple submodels via gating functions to achieve greater performance in numerous regression and classification applications. From a theoretical perspective, while there have been…

机器学习 · 统计学 2024-06-25 Huy Nguyen , Pedram Akbarian , TrungTin Nguyen , Nhat Ho

The Mixture of Experts (MoE) for language models has been proven effective in augmenting the capacity of models by dynamically routing each input token to a specific subset of experts for processing. Despite the success, most existing…

机器学习 · 计算机科学 2024-07-26 Hao Zhao , Zihan Qiu , Huijia Wu , Zili Wang , Zhaofeng He , Jie Fu

Multimodal large language models (MLLMs) have garnered widespread attention from researchers due to their remarkable understanding and generation capabilities in visual language tasks (e.g., visual question answering). However, the rapid…

计算机视觉与模式识别 · 计算机科学 2025-03-04 Tianyu Huai , Jie Zhou , Xingjiao Wu , Qin Chen , Qingchun Bai , Ze Zhou , Liang He

Large language models (LLM) have been attracting much attention from the community recently, due to their remarkable performance in all kinds of downstream tasks. According to the well-known scaling law, scaling up a dense LLM enhances its…

机器学习 · 计算机科学 2025-02-19 Zhenpeng Su , Xing Wu , Zijia Lin , Yizhe Xiong , Minxuan Lv , Guangyuan Ma , Hui Chen , Songlin Hu , Guiguang Ding

Continual learning can empower vision-language models to continuously acquire new knowledge, without the need for access to the entire historical dataset. However, mitigating the performance degradation in large-scale models is non-trivial…

计算机视觉与模式识别 · 计算机科学 2024-06-04 Jiazuo Yu , Yunzhi Zhuge , Lu Zhang , Ping Hu , Dong Wang , Huchuan Lu , You He

Mixture-of-Experts (MoE) has emerged as a promising approach to scale up deep learning models due to its significant reduction in computational resources. However, the dynamic nature of MoE leads to load imbalance among experts, severely…

分布式、并行与集群计算 · 计算机科学 2026-01-16 Chenqi Zhao , Wenfei Wu , Linhai Song , Yuchen Xu , Yitao Yuan

Mixture-of-Experts (MoE) models have gained popularity as a means of scaling the capacity of large language models (LLMs) while maintaining sparse activations and reduced per-token compute. However, in memory-constrained inference settings,…

机器学习 · 计算机科学 2026-03-23 Vivan Madan , Prajwal Singhania , Abhinav Bhatele , Tom Goldstein , Ashwinee Panda

Mixture-of-Experts (MoE) models improve the scalability of large language models (LLMs) by activating only a small subset of relevant experts per input. However, the sheer number of expert networks in an MoE model introduces a significant…

机器学习 · 计算机科学 2026-03-03 Qian Chen , Xianhao Chen , Kaibin Huang

Mixture of Experts (MoE) models enable parameter-efficient scaling through sparse expert activations, yet optimizing their inference and memory costs remains challenging due to limited understanding of their specialization behavior. We…

机器学习 · 计算机科学 2026-03-09 Marmik Chaudhari , Idhant Gulati , Nishkal Hundia , Pranav Karra , Shivam Raval

Continually expanding new languages for existing large language models (LLMs) is a promising yet challenging approach to building powerful multilingual LLMs. The biggest challenge is to make the model continuously learn new languages while…

计算与语言 · 计算机科学 2025-05-29 Xue Zhang , Yunlong Liang , Fandong Meng , Songming Zhang , Yufeng Chen , Jinan Xu , Jie Zhou

Graph incremental learning is a learning paradigm that aims to adapt trained models to continuously incremented graphs and data over time without the need for retraining on the full dataset. However, regular graph machine learning methods…

机器学习 · 计算机科学 2025-08-14 Lecheng Kong , Theodore Vasiloudis , Seongjun Yun , Han Xie , Xiang Song