CoLD: Counterfactually-Guided Length Debiasing for Process Reward Models in Mathematical Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zheng, Congmin, Zhu, Jiachen, Lin, Jianghao, Dai, Xinyi, Liu, Weiwen, Li, Haoxuan, Yu, Yong, Zhang, Weinan, Yang, Mengyue
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/2507.15698