Detect-Localize-Repair: A Unified Framework for Learning to Debug with CodeT5

TL;DR

This paper introduces CodeT5-DLR, a unified framework that leverages a pretrained language model to detect, localize, and repair bugs in code, demonstrating superior performance on Java and Python debugging datasets.

Contribution

It presents the first unified model for bug detection, localization, and repair, integrating these tasks into a single framework based on CodeT5.

Findings

Outperforms existing baselines in bug detection, localization, and repair.

Effective on datasets in both Java and Python.

Demonstrates the mutual benefits of joint debugging tasks.

Abstract

Automated software debugging is a crucial task for improving the productivity of software developers. Many neural-based techniques have been proven effective for debugging-related tasks such as bug localization and program repair (or bug fixing). However, these techniques often focus only on either one of them or approach them in a stage-wise manner, ignoring the mutual benefits between them. In this work, we propose a novel unified \emph{Detect-Localize-Repair} framework based on a pretrained programming language model CodeT5 to seamlessly address these tasks, named CodeT5-DLR. Specifically, we propose three objectives to adapt the generic CodeT5 for debugging: a bug detection objective to determine whether a given code snippet is buggy or not, a bug localization objective to identify the buggy lines, and a program repair objective to translate the buggy code to its fixed version. We evaluate it on each of these tasks and their combined setting on two newly collected line-level debugging datasets in Java and Python. Extensive results show that our model significantly outperforms existing baselines from both NLP and software engineering domains.