ReGVD: Revisiting Graph Neural Networks for Vulnerability Detection
This work addresses the challenge of automated vulnerability detection in software systems, which is crucial for cybersecurity but requires specialized expertise, though it appears incremental as it builds on existing GNN and pre-trained language model approaches.
The authors tackled the problem of identifying vulnerabilities in source code by proposing ReGVD, a graph neural network-based model that treats vulnerability detection as an inductive text classification problem, and it achieved the highest accuracy on the CodeXGLUE benchmark dataset.
Identifying vulnerabilities in the source code is essential to protect the software systems from cyber security attacks. It, however, is also a challenging step that requires specialized expertise in security and code representation. To this end, we aim to develop a general, practical, and programming language-independent model capable of running on various source codes and libraries without difficulty. Therefore, we consider vulnerability detection as an inductive text classification problem and propose ReGVD, a simple yet effective graph neural network-based model for the problem. In particular, ReGVD views each raw source code as a flat sequence of tokens to build a graph, wherein node features are initialized by only the token embedding layer of a pre-trained programming language (PL) model. ReGVD then leverages residual connection among GNN layers and examines a mixture of graph-level sum and max poolings to return a graph embedding for the source code. ReGVD outperforms the existing state-of-the-art models and obtains the highest accuracy on the real-world benchmark dataset from CodeXGLUE for vulnerability detection. Our code is available at: \url{https://github.com/daiquocnguyen/GNN-ReGVD}.