L. Hermansson

EPIG-Tree: Compute-Optimal Branching for Gradient-Efficient Reinforcement Learning

Reward-based reinforcement learning for language models, exemplified by Group Relative Policy Optimization (GRPO), collapses an entire stochastic trajectory into a single scalar reward. This is clean and scalable, but it explores and allocates reward inefficiently: a trajectory may contain many causal decisions, recove...