Hugging Face's logo
RLHFlow 's Collections
RLHFlow MATH Process Reward Model
Standard-format-preference-dataset
Mixture-of-preference-reward-modeling
RM-Bradley-Terry
PM-pair
Online RLHF
RLHFLow Reward Models
SFT Models

RM-Bradley-Terry

We train the reward model as the maximum likelihood estimation of the Bradley-Terry model.

Upvote
-