LProtector: An LLM-driven Vulnerability Detection System

TL;DR

Experiments on the Big-Vul dataset show that LProtector outperforms two state-of-the-art baselines in terms of F1 score, demonstrating the potential of integrating LLMs with vulnerability detection.

Abstract

This paper presents LProtector, an automated vulnerability detection system for C/C++ codebases driven by the large language model (LLM) GPT-4o and Retrieval-Augmented Generation (RAG). As software complexity grows, traditional methods face challenges in detecting vulnerabilities effectively. LProtector leverages GPT-4o's powerful code comprehension and generation capabilities to perform binary classification and identify vulnerabilities within target codebases. We conducted experiments on the Big-Vul dataset, showing that LProtector outperforms two state-of-the-art baselines in terms of F1 score, demonstrating the potential of integrating LLMs with vulnerability detection.

Figures (4)

• Figure 1: Architecture of LProtector
• Figure 2: Architecture of RAG
• Figure 3: Prompt without CoT
• Figure 4: Prompt with CoT