CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models
By: Hong Yi Lin , Chunhua Liu , Haoyu Gao and more
Potential Business Impact:
Helps computers fix code based on human notes.
State-of-the-art large language models (LLMs) have demonstrated impressive code generation capabilities but struggle with real-world software engineering tasks, such as revising source code to address code reviews, hindering their practical use. Code review comments are often implicit, ambiguous, and colloquial, requiring models to grasp both code and human intent. This challenge calls for evaluating large language models' ability to bridge both technical and conversational contexts. While existing work has employed the automated code refinement (ACR) task to resolve these comments, current evaluation methods fall short, relying on text matching metrics that provide limited insight into model failures and remain susceptible to training data contamination. To address these limitations, we introduce a novel evaluation benchmark, $\textbf{CodeReviewQA}$ that enables us to conduct fine-grained assessment of model capabilities and mitigate data contamination risks. In CodeReviewQA, we decompose the generation task of code refinement into $\textbf{three essential reasoning steps}$: $\textit{change type recognition}$ (CTR), $\textit{change localisation}$ (CL), and $\textit{solution identification}$ (SI). Each step is reformulated as multiple-choice questions with varied difficulty levels, enabling precise assessment of model capabilities, while mitigating data contamination risks. Our comprehensive evaluation spans 72 recently released large language models on $\textbf{900 manually curated, high-quality examples}$ across nine programming languages. Our results show that CodeReviewQA is able to expose specific model weaknesses in code review comprehension, disentangled from their generative automated code refinement results.
Similar Papers
Automated Code Review Using Large Language Models with Symbolic Reasoning
Software Engineering
AI finds coding mistakes better by thinking logically.
LAURA: Enhancing Code Review Generation with Context-Enriched Retrieval-Augmented LLM
Software Engineering
Helps computers write better code suggestions.
BitsAI-CR: Automated Code Review via LLM in Practice
Software Engineering
Helps computers find mistakes in code faster.