paperMay 2026Unreviewed

FuzzingBrain V2: A Multi-Agent LLM System for Automated Vulnerability Discovery and Reproduction

Ze Sheng, Zhicheng Chen, Qingxiao Xu, Kewen Zhu, Jeff Huang

Abstract

Software vulnerabilities pose critical security threats, with nearly 50,000 CVEs reported in 2025. While Large Language Models (LLMs) show promise for automated vulnerability detection, three key challenges remain. First, LLM-generated vulnerability reports suffer from high false positive rates and lack reproducible verification. Second, existing LLM-based approaches use suboptimal granularities for vulnerability localization: function-level analysis overlooks bugs when context becomes extensive

Cite

@misc{sheng2026fuzzingbrain,
  title = {{FuzzingBrain V2: A Multi-Agent LLM System for Automated Vulnerability Discovery and Reproduction}},
  author = {Ze Sheng and Zhicheng Chen and Qingxiao Xu and Kewen Zhu and Jeff Huang},
  year = {2026},
  month = may,
  eprint = {2605.21779},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.21779}
}