$\mathcal{B}$-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis

arXiv:2310.03173v2 Announce Type: replace
Abstract: Program synthesis aims to create accurate, executable programs from problem specifications, specifically from natural language descriptions in our context. Recent studies have leveraged the power of reinforcement learning (RL) in conjunction with large language models (LLMs), significantly enhancing code generation capabilities. The application of RL focuses on directly optimizing for functional correctness, offering an advantage over conventional supervised methods. Despite policy-based RL methods dominating the literature on RL for program synthesis, the nature of program …

abstract application arxiv capabilities code code generation coder context cs.cl language language models large language large language models llms natural natural language power reinforcement reinforcement learning studies synthesis type value