Enabling Automatic Repair of Source Code Vulnerabilities Using Data-Driven Methods

TL;DR

This paper explores enhancing data-driven models for automatic source code vulnerability repair by improving code representations, aiming to advance the effectiveness of fixing security flaws in software systems.

Contribution

It proposes new methods to improve code representations for vulnerability repair from data type, modeling, and task perspectives, addressing a gap in automatic security vulnerability fixing.

Findings

Enhanced code representations lead to better vulnerability repair performance.

Improved models show promise in fixing security vulnerabilities.

The approach advances automatic program repair techniques.

Abstract

Users around the world rely on software-intensive systems in their day-to-day activities. These systems regularly contain bugs and security vulnerabilities. To facilitate bug fixing, data-driven models of automatic program repair use pairs of buggy and fixed code to learn transformations that fix errors in code. However, automatic repair of security vulnerabilities remains under-explored. In this work, we propose ways to improve code representations for vulnerability repair from three perspectives: input data type, data-driven models, and downstream tasks. The expected results of this work are improved code representations for automatic program repair and, specifically, fixing security vulnerabilities.