Code Generation & Program Synthesis with RL Reading List

Curated by Mouhssine Rifaki | Stanford Electrical Engineering | Last updated August 2026

From Codex to AlphaCode and beyond, code generation now pairs large pretrained models with execution feedback and reinforcement learning. These papers chart that progression.

Code Generation & Program Synthesis with RL: 10 key papers

1. Evaluating Large Language Models Trained on Code
   Chen et al. arXiv 2021.
2. Program Synthesis with Large Language Models
   Austin et al. arXiv 2021.
3. Competition-Level Code Generation with AlphaCode
   Li et al. arXiv 2022.
4. Code Llama: Open Foundation Models for Code
   Rozière et al. arXiv 2023.
5. CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning
   Le et al. arXiv 2022.
6. CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis
   Nijkamp et al. arXiv 2022.
7. StarCoder: may the source be with you!
   Li et al. arXiv 2023.
8. InCoder: A Generative Model for Code Infilling and Synthesis
   Fried et al. arXiv 2022.
9. SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
   Jimenez et al. arXiv 2023.
10. RLTF: Reinforcement Learning from Unit Test Feedback
    Liu et al. arXiv 2023.

Related reading lists

• RL from Human Feedback (RLHF)
• Retrieval-Augmented Generation
• Diffusion Models & Score-Based Generative Models
• LLM Agents & Tool-Augmented Language Models
• Diffusion Models for Decision Making
• Neuroscience-Inspired Reinforcement Learning
• Instruction Following & Finetuning
• Curriculum Learning & Teacher-Student