Bridging the Gap between User Intent and LLM: A Requirement Alignment Approach for Code Generation

TL;DR

This paper introduces REA-Coder, a requirement alignment method that iteratively improves code generation accuracy from large language models by ensuring requirements are correctly understood and aligned.

Contribution

The paper presents a novel requirement alignment approach that enhances LLM-based code generation by iteratively aligning requirements and generated code, outperforming existing methods.

Findings

REA-Coder outperforms all baselines on five benchmarks.

Achieves average improvements of up to 30.25%.

Iterative requirement alignment improves code correctness.

Abstract

Code generation refers to automatically producing executable programs from user requirements. Recently, researchers have explored approaches to enhance the correctness of generated code with advanced large language models. Although achieving improvements, existing approaches focus on designing reasoning strategies or post-refinement methods to enhance code generation performance. Despite their differences, all these methods share a common assumption: the LLM can correctly understand the given requirement. However, this assumption does not always hold. To fill this gap, we propose REA-Coder, a requirement alignment approach to enhance the code generation performance of LLMs. REA-Coder involves first identifying the requirement content that does not align with LLMs and aligning the requirements. Then, based on the aligned requirements, LLMs generate code and further verify whether the generated code aligns with the requirements, iterating this process of requirement alignment and code generation until generating correct code or achieving the maximum number of iterations. Experimental results show that REA-Coder outperforms all advanced baselines on four LLMs across five programming benchmarks. Concretely, REA-Coder achieves average improvements of 7.93%, 30.25%, 26.75%, 8.59%, and 8.64% on the five benchmark datasets, demonstrating the effectiveness of requirement alignment for improving the code generation performance of LLMs.