paper-with-me

홈 › Papers

Adaptive Gating in Mixture-of-Experts based Language Models

2023-10-11 · Jiamin Li, Qiang Su, Yitao Yang, Yimin Jiang, Cong Wang, Hong Xu

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 while maintaining a constant number of computational operations. Existing MoE model adopts a fixed gating network where each token is computed by the same number of experts. However, this approach contradicts our intuition that the tokens in each sequence vary in terms of their linguistic complexity and, consequently, require different computational costs. Little is discussed in prior research on the trade-off between computation per token and model performance. This paper introduces adaptive gating in MoE, a flexible training strategy that allows tokens to be processed by a variable number of experts based on expert probability distribution. The proposed framework preserves sparsity while improving training efficiency. Additionally, curriculum learning is leveraged to further reduce training time. Extensive experiments on diverse NLP tasks show that adaptive gating reduces at most 22.5% training time while maintaining inference quality. Moreover, we conduct a comprehensive analysis of the routing decisions and present our insights when adaptive gating is used.

📄 PDF Abstract BibTeX arXiv:2310.07188

Code (0)

등록된 구현이 없습니다.

Tasks

Mixture-of-Experts

Similar Papers 제목 키워드 기반

Horseshoe Mixtures-of-Experts (HS-MoE)

2026-01-14 · Nick Polson, Vadim Sokolov arxiv

Horseshoe mixtures-of-experts (HS-MoE) models provide a Bayesian framework for sparse expert selection in mixture-of-experts architectures. We combine the horseshoe prior's adaptive global-local shrinkage with input-depe…

AdaMoLE: Fine-Tuning Large Language Models with Adaptive Mixture of Low-Rank Adaptation Experts

2024-05-01 · Zefang Liu, Jiahua Luo

We introduce AdaMoLE, a novel method for fine-tuning large language models (LLMs) through an Adaptive Mixture of Low-Rank Adaptation (LoRA) Experts. Moving beyond conventional methods that employ a static top-k strategy …

Multiple Choice Question Answering (MCQA)

Dynamic Mixture of Experts: An Auto-Tuning Approach for Efficient Transformer Models

2024-05-23 · Yongxin Guo, Zhenglin Cheng, Xiaoying Tang, Zhaopeng Tu 외

The Sparse Mixture of Experts (SMoE) has been widely employed to enhance the efficiency of training and inference for Transformer-based foundational models, yielding promising results. However, the performance of SMoE he…

Mixture-of-ExpertsVisual Question Answering

Convergence Rates for Softmax Gating Mixture of Experts

2025-03-05 · Huy Nguyen, Nhat Ho, Alessandro Rinaldo

Mixture of experts (MoE) has recently emerged as an effective framework to advance the efficiency and scalability of machine learning models by softly dividing complex tasks among multiple specialized sub-models termed e…

Mixture-of-Expertsparameter estimation

Little By Little: Continual Learning via Self-Activated Sparse Mixture-of-Rank Adaptive Learning

2025-06-26 · Haodong Lu, Chongyang Zhao, Jason Xue, Lina Yao 외

Continual learning (CL) with large pre-trained models is challenged by catastrophic forgetting and task interference. Existing LoRA-based Mixture-of-Experts (MoE) approaches mitigate forgetting by assigning and freezing …

Continual LearningMixture-of-Experts