Abstract:During software development and maintenance, vulnerability detection is an essential part of software quality assurance. Even though many program-analysis-based and machine-learning-based approaches have been proposed to automatically detect vulnerabilities, they rely on explicit rules or patterns defined by security experts and suffer from either high false positives or high false negatives. Recently, an increasing number of studies leverage deep learning techniques, especially Graph Neural Network (GNN), to detect vulnerabilities. These approaches leverage program analysis to represent the program semantics as graphs and perform graph analysis to detect vulnerabilities. However, they suffer from two main problems: (i) Existing GNN-based techniques do not effectively learn the structural and semantic features from source code for vulnerability detection. (ii) These approaches tend to ignore fine-grained information in source code. To tackle these problems, in this paper, we propose a novel vulnerability detection approach, named MGVD (M ultiple-G raph-Based V ulnerability D etection), to detect vulnerable functions. To effectively learn the structural and semantic features from source code, MGVD uses three different ways to represent each function into multiple forms, i.e., two statement graphs and a sequence of tokens. Then we encode such representations to a three-channel feature matrix. The feature matrix contains the structural feature and the semantic feature of the function. And we add a weight allocation layer to distribute the weights between structural and semantic features. To overcome the second problem, MGVD constructs each graph representation of the input function using multiple different graphs instead of a single graph. Each graph focuses on one statement in the function and its nodes denote the related statements and their fine-grained code elements. Finally, MGVD leverages CNN to identify whether this function is vulnerable based on such feature matrix. We conduct experiments on 3 vulnerability datasets with a total of 30,341 vulnerable functions and 127,931 non-vulnerable functions. The experimental results show that our method outperforms the state-of-the-art by 9.68% – 10.28% in terms of F1-score.

Learning to Locate and Describe Vulnerabilities

Function-Level Vulnerability Detection Through Fusing Multi-Modal Knowledge

Ignnvd: A Novel Software Vulnerability Detection Model Based on Integrated Graph Neural Networks

Vul-LMGNNs: Fusing Language Models and Online-Distilled Graph Neural Networks for Code Vulnerability Detection

Vulnerability Detection via Multiple-Graph-Based Code Representation

LineVD: Statement-level Vulnerability Detection using Graph Neural Networks

The Vulnerability Is in the Details: Locating Fine-grained Information of Vulnerable Code Identified by Graph-based Detectors

VulANalyzeR: Explainable Binary Vulnerability Detection with Multi-task Learning and Attentional Graph Convolution

VulEye: A Novel Graph Neural Network Vulnerability Detection Approach for PHP Application

Learning to map source code to software vulnerability using code-as-a-graph

VulGraB: Graph‐embedding‐based code vulnerability detection with bi‐directional gated graph neural network

Software Vulnerability Detection via Deep Learning over Disaggregated Code Graph Representation

Keep It Simple: Towards Accurate Vulnerability Detection for Large Code Graphs

VulDeeLocator: A Deep Learning-based Fine-grained Vulnerability Detector

TACSan: Enhancing Vulnerability Detection with Graph Neural Network

HyVulDect: A Hybrid Semantic Vulnerability Mining System Based on Graph Neural Network

Automatic CWE Assignment for Vulnerabilities with Graph Neural Networks

VulCNN

Vulnerability Detection by Learning from Syntax-Based Execution Paths of Code

DeepVulSeeker: A novel vulnerability identification framework via code graph structure and pre-training mechanism

Graph Neural Network Based Approach to Automatically Assigning Common Weakness Enumeration Identifiers for Vulnerabilities