HogVul: Black-box Adversarial Code Generation Framework Against LM-based Vulnerability Detectors

TL;DR

HogVul is a novel black-box adversarial framework that combines lexical and syntax perturbations using Particle Swarm Optimization to effectively evade LM-based vulnerability detectors, significantly improving attack success rates.

Contribution

This work introduces HogVul, a unified dual-channel optimization framework that enhances black-box adversarial code generation against vulnerability detectors by integrating lexical and syntax perturbations.

Findings

Achieves 26.05% higher attack success rate than baselines

Effectively explores a larger adversarial code space

Demonstrates the effectiveness of hybrid optimization strategies

Abstract

Recent advances in software vulnerability detection have been driven by Language Model (LM)-based approaches. However, these models remain vulnerable to adversarial attacks that exploit lexical and syntax perturbations, allowing critical flaws to evade detection. Existing black-box attacks on LM-based vulnerability detectors primarily rely on isolated perturbation strategies, limiting their ability to efficiently explore the adversarial code space for optimal perturbations. To bridge this gap, we propose HogVul, a black-box adversarial code generation framework that integrates both lexical and syntax perturbations under a unified dual-channel optimization strategy driven by Particle Swarm Optimization (PSO). By systematically coordinating two-level perturbations, HogVul effectively expands the search space for adversarial examples, enhancing the attack efficacy. Extensive experiments on four benchmark datasets demonstrate that HogVul achieves an average attack success rate improvement of 26.05\% over state-of-the-art baseline methods. These findings highlight the potential of hybrid optimization strategies in exposing model vulnerabilities.