CURE: Code-Aware Neural Machine Translation for Automatic Program Repair

TL;DR

CURE introduces a code-aware neural machine translation approach for automatic program repair, leveraging pre-training, a specialized search strategy, and subword tokenization to improve fix accuracy over existing methods.

Contribution

CURE's novel combination of pre-trained programming language models, code-aware search, and subword tokenization advances neural APR beyond prior template-based and NMT approaches.

Findings

Correctly fixed 57 Defects4J bugs

Fixed 26 QuixBugs bugs

Outperformed all existing APR techniques on benchmarks

Abstract

Automatic program repair (APR) is crucial to improve software reliability. Recently, neural machine translation (NMT) techniques have been used to fix software bugs automatically. While promising, these approaches have two major limitations. Their search space often does not contain the correct fix, and their search strategy ignores software knowledge such as strict code syntax. Due to these limitations, existing NMT-based techniques underperform the best template-based approaches. We propose CURE, a new NMT-based APR technique with three major novelties. First, CURE pre-trains a programming language (PL) model on a large software codebase to learn developer-like source code before the APR task. Second, CURE designs a new code-aware search strategy that finds more correct fixes by focusing on compilable patches and patches that are close in length to the buggy code. Finally, CURE uses a subword tokenization technique to generate a smaller search space that contains more correct fixes. Our evaluation on two widely-used benchmarks shows that CURE correctly fixes 57 Defects4J bugs and 26 QuixBugs bugs, outperforming all existing APR techniques on both benchmarks.