Jointly Learning to Repair Code and Generate Commit Message

TL;DR

This paper introduces a unified approach for simultaneously repairing buggy code and generating commit messages, leveraging a joint model and enhanced training methods to improve performance in both tasks.

Contribution

It presents the first joint model for code repair and commit message generation, along with a multilingual dataset and enhanced training strategies to address error propagation.

Findings

Joint model outperforms cascaded models in commit message generation.

Enhanced cascaded models with teacher-student and multi-task methods improve code repair accuracy.

The joint model effectively reduces error propagation between tasks.

Abstract

We propose a novel task of jointly repairing program codes and generating commit messages. Code repair and commit message generation are two essential and related tasks for software development. However, existing work usually performs the two tasks independently. We construct a multilingual triple dataset including buggy code, fixed code, and commit messages for this novel task. We provide the cascaded models as baseline, which are enhanced with different training approaches, including the teacher-student method, the multi-task method, and the back-translation method. To deal with the error propagation problem of the cascaded method, the joint model is proposed that can both repair the code and generate the commit message in a unified framework. Experimental results show that the enhanced cascaded model with teacher-student method and multitask-learning method achieves the best score on different metrics of automated code repair, and the joint model behaves better than the cascaded model on commit message generation.