Litcius/Paper detail

Vul-RAG: Enhancing LLM-based Vulnerability Detection via Knowledge-level RAG

Xueying Du, Geng Zheng, Kaixin Wang, Yi Zou, Yujia Wang, Wentai Deng, Jiayi Feng, Mingwei Liu, Bihuan Chen, Xin Peng, Tao Ma, Yiling Lou

2026ACM Transactions on Software Engineering and Methodology7 citationsDOI

Abstract

Although LLMs have shown promising potential in vulnerability detection, this study reveals their limitations in distinguishing between vulnerable and similar-but-benign patched code (only 0.06 - 0.14 accuracy). It shows that LLMs struggle to capture the root causes of vulnerabilities during vulnerability detection. To address this challenge, we propose enhancing LLMs with multi-dimensional vulnerability knowledge distilled from historical vulnerabilities and fixes. We design a novel knowledge-level Retrieval-Augmented Generation framework Vul-RAG, which improves LLMs with an accuracy increase of 16% - 24% in identifying vulnerable and patched code. Additionally, vulnerability knowledge generated by Vul-RAG can further (1) serve as high-quality explanations to improve manual detection accuracy (from 60% to 77%), and (2) detect 10 previously-unknown bugs in the recent Linux kernel release with 6 assigned CVEs.

Topics & Concepts

Vulnerability (computing)Computer scienceComputer securityCode (set theory)Risk analysis (engineering)Risk assessmentRisk managementVulnerability assessmentRoot causeInternet privacyAdvanced Malware Detection TechniquesSecurity and Verification in ComputingSoftware Engineering Research
Vul-RAG: Enhancing LLM-based Vulnerability Detection via Knowledge-level RAG | Litcius