EDGC: Entropy-driven Dynamic Gradient Compression for Efficient LLM Training

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yi, Qingao, Duan, Jiaang, Hu, Hanwen, Hua, Qin, Zhao, Haiyan, Qian, Shiyou, Yang, Dingyu, Cao, Jian, Tang, Jinghua, Yu, Yinghao, Liao, Chenzhi, Wang, Kangjin, Zhang, Liping
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence Performance
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2511.10333