arXiv · 2501.15316
ToMoE: Converting Dense Large Language Models to Mixture-of-Experts through Dynamic Structural Pruning
Abstract
Large Language Models (LLMs) have demonstrated remarkable abilities in tackling a wide range of complex tasks. However, their huge computational and memory costs raise significant challenges in deploying these models on resource-constrained devices or efficiently serving them. Prior approaches have attempted to alleviate these problems by permanently removing less important model structures, yet these methods often result in substantial performance degradation due to the permanent deletion of model parameters. In this work, we tried to mitigate this issue by reducing the number of active parameters without permanently removing them. Specifically, we introduce a differentiable dynamic pruning method that pushes dense models to maintain a fixed number of active parameters by converting their MLP layers into a Mixture of Experts (MoE) architecture. Our method, even without fine-tuning, consistently outperforms previous structural pruning techniques across diverse model families, including Phi-2, LLaMA-2, LLaMA-3, and Qwen-2.5.
Explore related subjects
Keep this discovery
Shangqian Gao, Ting Hua, Reza Shirkavand, Chi-Heng Lin, Zheng Tang, Zhengao Li, Longge Yuan, Fangyi Li, Zeyu Zhang, Alireza Ganjdanesh, Lou Qian, Xu Jie, Yen-Chang Hsu. 2025-01-25. ToMoE: Converting Dense Large Language Models to Mixture-of-Experts through Dynamic Structural Pruning. https://arxiv.org/abs/2501.15316
Cite the original work for its findings. Save a collection to share your selection of sources.