Interpreting Language Model Preferences Through the Lens of Decision Trees
A decision-tree perspective to interpret LLM preference mechanisms.
Posts
Interpretable Preferences via Multi-Objective Reward Modeling and Mixture-of-Experts
An interpretable reward modeling approach.
Alignment Guidebook
A guidebook for LLM alignment.
Reward Modeling Part 1: Bradley-Terry Model
This is the recipe for the RLHFlow/RLHF-Reward-Modeling repository used to train the reward model for RLHF.