MoEQuant: Enhancing Quantization for Mixture-of-Experts Large Language Models via Expert-Balanced Sampling and Affinity Guidance

Zhixuan Chen; Xing Hu; Dawei Yang; Zukang Xu; Xu Chen; Zhihang Yuan; Sifan Zhou; Jiangyong Yu

2025 ICML ICML 2025

MoEQuant: Enhancing Quantization for Mixture-of-Experts Large Language Models via Expert-Balanced Sampling and Affinity Guidance

Abstract

No abstract available. No abstract was captured for this paper when its proceedings were scraped - most often an older edition published without one online. How the corpus is built.

Authors

Zhixuan Chen, Xing Hu, Dawei Yang, Zukang Xu, Xu Chen, Zhihang Yuan, Sifan Zhou, Jiangyong Yu

Topics

Machine Learning > Application Areas > Efficient Computing Artificial Intelligence > Core AI > Large Language Models

Keywords

quantization dynamic routing sparse activation mixture-of-experts affinity guidance expert-balanced sampling

Download PDF