Learning Less Is More: Premature Upper-Layer Attention Specialization Hurts Language Model Pretraining

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhu, Jinchang, Li, Jindong, Hao, Yuwen, Zou, Chengyu, Fu, Rong, Yang, Menglin
Format:	Preprint
Published:	2026
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/2605.10504