StaRPO: Stability-Augmented Reinforcement Policy Optimization
StaRPO is a reinforcement learning framework designed to improve how large language models reason. It addresses the flaw in existing methods that only reward correct final answers, leading to inconsistent reasoning chains. StaRPO introduces two stability metrics—Autocorrelation Function (ACF) for local consistency and Path Efficiency (PE) for global goal-directedness—to create a new reward system that simultaneously boosts both the accuracy and logical soundness of the model's reasoning process.