HierMoE: Accelerating MoE Training with Hierarchical Token Deduplication and Expert Swap

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Lin, Wenxiang, Pan, Xinglin, Zhang, Lin, Shi, Shaohuai, Wang, Xuan, Chu, Xiaowen
Format:	Preprint
Published:	2025
Subjects:	Distributed, Parallel, and Cluster Computing Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

Similar Items