Reinforcement Learning for Large Language Model Fine-Tuning: A Systematic Literature Review

Large Language Models (LLMs) have been developed for a wide range of language-based tasks, while Reinforcement Learning (RL) has been primarily applied to decision-making problems such as robotics, game theory, and control systems. Nowadays, these two paradigms are integrated through different synergies. In this literature review, we focus on \textit{RL4LLM fine-tuning}, where RL techniques are systematically leveraged to fine-tune LLMs and align them with various preferences. Our review provides a comprehensive analysis of 230 recent publications, presenting a methodological taxonomy that organizes current research into three primary method domains: \textit{Optimization Algorithm}, concerning innovation in core RL update rules; \textit{Training Framework}, regarding innovation in the orchestration of the training process; and \textit{Reward Modeling}, addressing how LLMs learn and represent preferences and feedback. Within these primary domains, we further analyze methods and innovations through more granular categories to provide an in-depth summary of RL4LLM fine-tuning research. We address three research questions: 1) recent methods overview, 2) methodological innovations, and 3) limitations and future directions. Our analysis comprehensively demonstrates the breadth and impact of recent RL4LLM fine-tuning research while highlighting valuable directions for future investigation.