VulnAgent-X: A Layered Agentic Framework for Repository-Level Vulnerability Detection
Renwei Meng ∗, Haoyi Wu ∗, Jingming Wang ∗, Haoyan Bai · 2026
Read VulnAgent-X: A Layered Agentic Framework for Repository-Level Vulnerability Detection by Renwei Meng ∗, Haoyi Wu ∗, Jingming Wang ∗, Haoyan Bai (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-13384.txt.
Description
Renwei Meng, Haoyi Wu, and Jingming Wang present VulnAgent-R2 for repository-level vulnerability detection. Flaws often hinge on cross-file data flow, build options, and runtime guards, so lone function checks give fragile warnings. Their budget-aware system adds counterfactual evidence reweighting, build-aware verification plans, and a cost-risk scheduler. Graph triage, role-specialized agents, and sceptic counter-evidence feed a calibrated fusion step. On four public datasets it beat prior agents on F1 and AUROC while using far fewer online tokens.
Catalog record
| CREATOR | Renwei Meng ∗, Haoyi Wu ∗, Jingming Wang ∗, Haoyan Bai |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 17168 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.13384