Adversarial Patch Generation for Automated Program Repair

TL;DR

This paper introduces NEVERMORE, a learning-based adversarial approach using GANs to generate automated program repairs that closely mimic human fixes, showing promising results on real-world bugs and benchmark datasets.

Contribution

NEVERMORE is the first GAN-based framework for automated program repair trained on historical fixes, improving repair accuracy and scalability over existing methods.

Findings

NEVERMORE matches human fixes for 21.2% of bugs in real-world evaluation.

Successfully repairs 4 bugs in Defects4J unresolved by other methods.

Fixes an additional 8 bugs only solved by some baseline approaches.

Abstract

Automated Program Repair has attracted significant research in recent years, leading to diverse techniques that focus on two main directions: search-based and semantic-based program repair. The former techniques often face challenges due to the vast search space, resulting in difficulties in identifying correct solutions, while the latter approaches are constrained by the capabilities of the underlying semantic analyser, limiting their scalability. In this paper, we propose NEVERMORE, a novel learning-based mechanism inspired by the adversarial nature of bugs and fixes. NEVERMORE is built upon the Generative Adversarial Networks architecture and trained on historical bug fixes to generate repairs that closely mimic human-produced fixes. Our empirical evaluation on 500 real-world bugs demonstrates the effectiveness of NEVERMORE in bug-fixing, generating repairs that match human fixes for 21.2% of the examined bugs. Moreover, we evaluate NEVERMORE on the Defects4J dataset, where our approach generates repairs for 4 bugs that remained unresolved by state-of-the-art baselines. NEVERMORE also fixes another 8 bugs which were only resolved by a subset of these baselines. Finally, we conduct an in-depth analysis of the impact of input and training styles on NEVERMORE's performance, revealing where the chosen style influences the model's bug-fixing capabilities.