ConfPO: Exploiting Policy Model Confidence for Critical Token Selection in Preference Optimization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yoon, Hee Suk, Yoon, Eunseop, Hasegawa-Johnson, Mark, Kim, Sungwoong, Yoo, Chang D.
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2506.08712