TY - GEN
T1 - LAURA
T2 - 2025 40th IEEE/ACM International Conference on Automated Software Engineering, ASE 2025
AU - Zhang, Yuxin
AU - Zhang, Yuxia
AU - Sun, Zeyu
AU - Jiang, Yanjie
AU - Liu, Hui
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Code review is critical for ensuring software quality and maintainability. With the rapid growth in software scale and complexity, code review has become a bottleneck in the development process because of its time-consuming and knowledge-intensive nature and the shortage of experienced developers willing to review code. Several approaches have been proposed for automatically generating code reviews based on retrieval, neural machine translation, pre-trained models, or large language models (LLMs). These approaches mainly leverage historical code changes and review comments. However, a large amount of crucial information for code review, such as the context of code changes and prior review knowledge, has been overlooked. This paper proposes an LLM-based review knowledge-augmented, context-aware framework for code review generation, named LAURA. The framework integrates review exemplar retrieval, context augmentation, and systematic guidance to enhance the performance of ChatGPT-4o and DeepSeek v3 in generating code review comments. Besides, given the extensive low-quality reviews in existing datasets, we also constructed a high-quality dataset. Experimental results show that for both models, LAURA generates review comments that are either completely correct or at least helpful to developers in 42.2% and 40.4% of cases, respectively, significantly outperforming SOTA baselines. Furthermore, our ablation studies demonstrate that all components of LAURA contribute positively to improving comment quality.
AB - Code review is critical for ensuring software quality and maintainability. With the rapid growth in software scale and complexity, code review has become a bottleneck in the development process because of its time-consuming and knowledge-intensive nature and the shortage of experienced developers willing to review code. Several approaches have been proposed for automatically generating code reviews based on retrieval, neural machine translation, pre-trained models, or large language models (LLMs). These approaches mainly leverage historical code changes and review comments. However, a large amount of crucial information for code review, such as the context of code changes and prior review knowledge, has been overlooked. This paper proposes an LLM-based review knowledge-augmented, context-aware framework for code review generation, named LAURA. The framework integrates review exemplar retrieval, context augmentation, and systematic guidance to enhance the performance of ChatGPT-4o and DeepSeek v3 in generating code review comments. Besides, given the extensive low-quality reviews in existing datasets, we also constructed a high-quality dataset. Experimental results show that for both models, LAURA generates review comments that are either completely correct or at least helpful to developers in 42.2% and 40.4% of cases, respectively, significantly outperforming SOTA baselines. Furthermore, our ablation studies demonstrate that all components of LAURA contribute positively to improving comment quality.
KW - Code Review Generation
KW - Context-aware
KW - LLMs
KW - Review Exemplar Retrieval
UR - https://www.scopus.com/pages/publications/105034688706
U2 - 10.1109/ASE63991.2025.00245
DO - 10.1109/ASE63991.2025.00245
M3 - Conference contribution
AN - SCOPUS:105034688706
T3 - Proceedings - 2025 40th IEEE/ACM International Conference on Automated Software Engineering, ASE 2025
SP - 2983
EP - 2995
BT - Proceedings - 2025 40th IEEE/ACM International Conference on Automated Software Engineering, ASE 2025
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 16 November 2025 through 20 November 2025
ER -