Self-Guided Process Reward Optimization with Redefined Step-wise Advantage for Process Reinforcement Learning

TL;DR

This paper introduces SPRO, a process reward optimization framework for Process Reinforcement Learning that derives intrinsic rewards and step-wise advantages, improving training efficiency and performance without extra computational costs.

Contribution

SPRO provides a unified theoretical framework for process-level advantage estimation and intrinsic reward derivation, enhancing efficiency and stability in process RL for LLMs.

Findings

SPRO achieves 3.4x higher training efficiency than vanilla GRPO.

SPRO improves test accuracy by 17.5%.

SPRO reduces response length by about one-third.

Abstract

Process Reinforcement Learning~(PRL) has demonstrated considerable potential in enhancing the reasoning capabilities of Large Language Models~(LLMs). However, introducing additional process reward models incurs substantial computational overhead, and there is no unified theoretical framework for process-level advantage estimation. To bridge this gap, we propose \textbf{S}elf-Guided \textbf{P}rocess \textbf{R}eward \textbf{O}ptimization~(\textbf{SPRO}), a novel framework that enables process-aware RL through two key innovations: (1) we first theoretically demonstrate that process rewards can be derived intrinsically from the policy model itself, and (2) we introduce well-defined cumulative process rewards and \textbf{M}asked \textbf{S}tep \textbf{A}dvantage (\textbf{MSA}), which facilitates rigorous step-wise action advantage estimation within shared-prompt sampling groups. Our experimental results demonstrate that SPRO outperforms vaniila GRPO with 3.4x higher training efficiency and a 17.5\% test accuracy improvement. Furthermore, SPRO maintains a stable and elevated policy entropy throughout training while reducing the average response length by approximately $1/3$, evidencing sufficient exploration and prevention of reward hacking. Notably, SPRO incurs no additional computational overhead compared to outcome-supervised RL methods such as GRPO, which benefit industrial implementation.