Probing Preference Representations: A Multi-Dimensional Evaluation and Analysis Method for Reward Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Chenglong, Huo, Yifu, Gan, Yang, Mu, Yongyu, He, Qiaozhi, Yang, Murun, Li, Bei, Zhang, Chunliang, Liu, Tongran, Ma, Anxiang, Yu, Zhengtao, Zhu, Jingbo, Xiao, Tong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!