Early Exit Is a Natural Capability in Transformer-based Models: An Empirical Study on Early Exit without Joint Optimization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Shan, Weiqiao, Meng, Long, Zheng, Tong, Luo, Yingfeng, Li, Bei, Wang, junxin, Xiao, Tong, Zhu, Jingbo
Format:	Preprint
Published:	2024
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2412.01455