Diversifying the Expert Knowledge for Task-Agnostic Pruning in Sparse Mixture-of-Experts

Zhang, Zeliang, Liu, Xiaodong, Cheng, Hao, Xu, Chenliang, Gao, Jianfeng

arXiv.org Artificial Intelligence 

By increasing model parameters but activating them sparsely when performing a task, the use of Mixture-of-Experts (MoE) architecture significantly improves the performance of Large Language Models (LLMs) without increasing the inference cost. However, the memory consumption due to the growing number of experts presents a challenge to the deployment of these models in many real world settings. Our empirical study reveals that some experts encode redundant knowledge during pre-training. We thus propose a method of grouping and pruning similar experts to improve model's parameter efficiency. We validate the effectiveness of our method by pruning two state-of-the-art MoE models, Mixtral-8x7B and Mixtral-8x22B. Evaluation shows that our method outperforms other model pruning methods on a range of natural language tasks. To facilitate future research, we will release our code and the pruned MoE models.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found