Training Language Models to Generate Quality Code with Program Analysis Feedback

Code generation with large language models (LLMs), often termed vibe coding, is increasingly adopted in production but fails to ensure code quality, particularly in security (e. g. , SQL injection vulnerabilities) and maintainability (e. g. , missing type annotations). Existing methods, such as supervised fine-tuning and rule-based post-processing, rely on labor-intensive annotations or brittle heuristics, limiting their scalability and effectiveness. We propose REAL (Reinforcement rEwards from Automated anaLysis), a reinforcement learning framework that trains LLMs to generate production-quality code using program analysis–guided feedback. Specifically, REAL integrates two automated signals: (1) static analyzers detecting security and maintainability defects and (2) unit tests ensuring functional correctness. Unlike prior work, our framework is prompt-agnostic and reference-free, enabling scalable supervision without manual intervention. Experiments across multiple datasets and model scales demonstrate that REAL outperforms state-of-the-art methods in simultaneous assessments of functionality and code quality. Our work bridges the gap between rapid prototyping and production-ready code, enabling LLMs to deliver both speed and quality.

Authors

• Feng Yao Tsinghua University
• Zilong Wang University of California, San Diego
• Liyuan Liu
• Junxia Cui
• Li Zhong University of California, San Diego
• Xiaohan Fu
• Haohui Mai
• Viswanathan Krishnan
• Jianfeng Gao
• Jingbo Shang University of California, San Diego

Keywords

No keywords are indexed for this paper.