This paper presents VLAI, a transformer-based model that predicts software vulnerability severity levels directly from text descriptions. Built on RoBERTa, VLAI is fine-tuned on over 600,000 real-world vulnerabilities and achieves over 82% accuracy in predicting severity categories, enabling faster and more consistent triage ahead of manual CVSS scoring. The model and dataset are open-source and integrated into the Vulnerability-Lookup service.
@article{arxiv.2507.03607,
title = {VLAI: A RoBERTa-Based Model for Automated Vulnerability Severity Classification},
author = {Cédric Bonhomme and Alexandre Dulaunoy},
journal= {arXiv preprint arXiv:2507.03607},
year = {2025}
}
Comments
This paper is a preprint for the 25V4C-TC: 2025 Vulnerability Forecasting Technical Colloquia. Darwin College Cambridge, UK, September 25-26, 2025