Online Pseudo-average Shifting Attention(PASA) for Robust Low-precision LLM Inference: Algorithms and Numerical Analysis

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Cheng, Long, Liao, Qichen, Wu, Fan, Mu, Junlin, Han, Tengfei, Qiu, Zhe, Li, Lianqiang, Liu, Tianyi, Miao, Fangzheng, Gao, Keming, Wang, Liang, Zhang, Zhen, Yin, Qiande
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence Numerical Analysis Performance
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2503.01873