Rewiring transformers for exploit likelihood prediction of cyber vulnerabilities
Document Type
Conference Proceeding
Source of Publication
Aip Conference Proceedings
Publication Date
5-21-2026
Abstract
This paper addresses the challenge of predicting the exploit likelihood of cyber vulnerabilities, a critical aspect of vulnerability management in the world of evolving global cyber threats. Traditional vulnerability scoring systems like CVSS and machine learning models struggle to accurately capture exploitability dynamics due to their resource-intensive nature and reliance on static datasets. Our proposed solution involves fine-tuning advanced transformer models, specifically DistilBERT and GPT2-Large, to predict exploitability vulnerabilities by integrating textual descriptions and numerical CVSS scores. This method demonstrates significant improvements in predicting accuracy and efficiency using the datasets from the National Vulnerability Database (NVD). The DistilBERT model outperformed the GPT2 model in this task. The study underscores the potential of transformer-based architectures to revolutionize vulnerability management processes, enabling organizations to allocate resources more effectively for risk mitigation.
DOI Link
ISSN
Publisher
AIP Publishing
Volume
3410
Issue
1
Disciplines
Computer Sciences
Keywords
cyber vulnerability, DistilBert, exploit likelihood, GPT2 transformer, large language models
Scopus ID
Recommended Citation
Kumar, Pranav; Agarwal, Harshit; Sharma, Vandana; Husain, Iqra; Hewage, Pradeep; and Iwendi, Celestine, "Rewiring transformers for exploit likelihood prediction of cyber vulnerabilities" (2026). All Works. 8126.
https://zuscholars.zu.ac.ae/works/8126
Indexed in Scopus
yes
Open Access
no