Fine-Tuning Open-Source Large Language Models for Cybersecurity and IT Support: A Comparative Study on LLM Fine-Tuning Techniques
IEEE Open Journal of the Computer Society · 2026 · 7 · 793-804
DOI: 10.1109/ojcs.2026.3685799 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Open Journal of the Computer Society, 2026. Verfasst von Chaithanya Vamshi Sai und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Birmingham City University and METCLOUD, Knowledge Transfer Partnership, Innovate U.K. Research and Innovation.
Abstract (Originalsprache)
The focus of Artificial Intelligence (AI) has progressed from exploring whether Large Language Models (LLMs) can transform domain-specific tasks to understanding how to optimise their impact. Advances in Natural Language Processing (NLP) have enabled LLMs to address complex tasks such as question answering, text classification, and translation with notable accuracy and efficiency. Hence, this study evaluates the effectiveness of fine-tuning open-source LLMs, which range from 2 to 9 billion parameters, for cybersecurity and IT support using three advanced techniques: Supervised Fine-Tuning (SFT), Low-Rank Adaptation (LoRA), and Quantised Low-Rank Adaptation (QLoRA). The results indicate that SFT achieves superior convergence in loss metrics for both tasks, while LoRA and QLoRA significantly reduce GPU memory and computational costs by 99% and 60%, respectively, with comparable performance. Models such as Llama, Mistral, and Phi consistently demonstrated high generalisation across all methods, making them particularly suitable for deployment in resource-constrained environments.
Förderung
- Birmingham City University and METCLOUD
- Knowledge Transfer Partnership
- Innovate U.K. Research and Innovation
Zitieren
APA
Sai, C. V., Elmitwally, N. S., Rice, I., Mahmoud, H., Vickers, I., & Schmoor, X. (2026). Fine-Tuning Open-Source Large Language Models for Cybersecurity and IT Support: A Comparative Study on LLM Fine-Tuning Techniques. IEEE Open Journal of the Computer Society, 7, 793-804. https://doi.org/10.1109/ojcs.2026.3685799
Vancouver
Sai CV, Elmitwally NS, Rice I, Mahmoud H, Vickers I, Schmoor X. Fine-Tuning Open-Source Large Language Models for Cybersecurity and IT Support: A Comparative Study on LLM Fine-Tuning Techniques. IEEE Open Journal of the Computer Society. 2026;7:793-804. doi:10.1109/ojcs.2026.3685799
BibTeX
@article{Sai2026,
author = {Sai, Chaithanya Vamshi and Elmitwally, Nouh Sabri and Rice, Iain and Mahmoud, Haitham and Vickers, Ian and Schmoor, Xavier},
title = {Fine-Tuning Open-Source Large Language Models for Cybersecurity and IT Support: A Comparative Study on LLM Fine-Tuning Techniques},
journal = {IEEE Open Journal of the Computer Society},
year = {2026},
volume = {7},
pages = {793--804},
doi = {10.1109/ojcs.2026.3685799}
}
EUVIMED