Iterative Refinement of Project-Level Code Context for Precise Code Generation with Compiler Feedback
This addresses the challenge of precise code generation for developers by integrating project-level context, though it is incremental as it builds on existing LLM and retrieval-based methods.
The paper tackles the problem of LLM-generated code containing errors due to missing project-specific context by introducing CoCoGen, an approach that uses compiler feedback and static analysis to iteratively refine code, resulting in over 80% improvement in generating context-dependent code compared to vanilla LLMs.
Large Language Models (LLMs) have shown remarkable progress in automated code generation. Yet, LLM-generated code may contain errors in API usage, class, data structure, or missing project-specific information. As much of this project-specific context cannot fit into the prompts of LLMs, we must find ways to allow the model to explore the project-level code context. We present CoCoGen, a new code generation approach that uses compiler feedback to improve the LLM-generated code. CoCoGen first leverages static analysis to identify mismatches between the generated code and the project's context. It then iteratively aligns and fixes the identified errors using information extracted from the code repository. We integrate CoCoGen with two representative LLMs, i.e., GPT-3.5-Turbo and Code Llama (13B), and apply it to Python code generation. Experimental results show that CoCoGen significantly improves the vanilla LLMs by over 80% in generating code dependent on the project context and consistently outperforms the existing retrieval-based code generation baselines.