M-ANT: Efficient Low-bit Group Quantization for LLMs via Mathematically Adaptive Numerical Type

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hu, Weiming, Zhang, Haoyan, Guo, Cong, Feng, Yu, Guan, Renyang, Hua, Zhendong, Liu, Zihan, Guan, Yue, Guo, Minyi, Leng, Jingwen
Format:	Preprint
Published:	2025
Subjects:	Hardware Architecture
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2502.18755