MoE-Prism: Disentangling Monolithic Experts for Elastic MoE Services via Model-System Co-Designs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xia, Xinfeng, Liu, Jiacheng, Hou, Xiaofeng, Tang, Peng, Zhang, Mingxuan, Wang, Wenfeng, Li, Chao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!