Skip to main content
Search

# Entropy-Guided Credit Assignment

1 related articles

Subscribe to updates
  1. arXiv

    EAPO couples entropy with the advantage sign for asymmetric credit assignment, topping average accuracy across four backbones

    The work introduces Entropic Advantage Policy Optimization (EAPO), which couples normalized policy entropy with the sign of the response advantage to reinforce high-entropy decisions in successful responses, penalize low-entropy decisions in failed responses, and attenuate penalties at uncertain positions, redistributing the response advantage to token level without auxiliary models, extra sampling, or privileged information, and achieving the best overall performance on mathematical, logical, and algorithmic reasoning tasks.

You’re all caught up