RACE: Retrieval-Augmented Commit Message Generation
This addresses the need for more accurate and less repetitive commit messages in software development, representing an incremental improvement over existing neural approaches.
The paper tackles the problem of repetitive or redundant commit messages in software development by proposing RACE, a retrieval-augmented neural method that uses retrieved similar commits as exemplars and includes an exemplar guider to handle inaccuracies, resulting in outperforming all baselines and boosting existing Seq2Seq models.
Commit messages are important for software development and maintenance. Many neural network-based approaches have been proposed and shown promising results on automatic commit message generation. However, the generated commit messages could be repetitive or redundant. In this paper, we propose RACE, a new retrieval-augmented neural commit message generation method, which treats the retrieved similar commit as an exemplar and leverages it to generate an accurate commit message. As the retrieved commit message may not always accurately describe the content/intent of the current code diff, we also propose an exemplar guider, which learns the semantic similarity between the retrieved and current code diff and then guides the generation of commit message based on the similarity. We conduct extensive experiments on a large public dataset with five programming languages. Experimental results show that RACE can outperform all baselines. Furthermore, RACE can boost the performance of existing Seq2Seq models in commit message generation.