RL-STaR: Theoretical Analysis of Reinforcement Learning Frameworks for Self-Taught Reasoner

The reasoning abilities of large language models (LLMs) have improved with chain-of-thought (CoT) prompting, allowing models to solve complex tasks stepwise. However, training CoT capabilities requires detailed reasoning data, which is often scarce. The self-taught reasoner (STaR) framework addresses this by using reinforcement learning to automatically generate reasoning steps, reducing reliance on human-labeled data. Although STaR and its variants have demonstrated empirical success, a theoretical foundation explaining these improvements is lacking. This work provides a theoretical framework for understanding the effectiveness of reinforcement learning on CoT reasoning and STaR. Our contributions are: (1) criteria for the quality of pre-trained models necessary to initiate effective reasoning improvement; (2) an analysis of policy improvement, showing why LLM reasoning improves iteratively with STaR; (3) conditions for convergence to an optimal reasoning policy; and (4) an examination of STaR's robustness, explaining how it can improve reasoning even when incorporating occasional incorrect steps; This framework aims to bridge empirical findings with theoretical insights, advancing reinforcement learning approaches for reasoning in LLMs.

@article{chang2025_2410.23912,
  title={ RL-STaR: Theoretical Analysis of Reinforcement Learning Frameworks for Self-Taught Reasoner },
  author={ Fu-Chieh Chang and Yu-Ting Lee and Hui-Ying Shih and Yi Hsuan Tseng and Pei-Yuan Wu },
  journal={arXiv preprint arXiv:2410.23912},
  year={ 2025 }
}