EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Small Language Models for Phishing Website Detection: Cost, Performance, and Privacy Trade-Offs

Georg Goldenits, Philip König, Sebastian Raubitzek, Andreas Ekelhart

Journal of Cybersecurity and Privacy · 05.03.2026 · 6 (2) · 48

DOI: 10.3390/jcp6020048 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Cybersecurity and Privacy, online erschienen am 05.03.2026. Verfasst von Georg Goldenits und 3 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Austrian Research Promotion Agency, BMIMI, BMWET.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Phishing websites pose a major cybersecurity threat, exploiting unsuspecting users and causing significant financial and organisational harm. Traditional machine learning approaches for phishing detection often require extensive feature engineering, continuous retraining, and costly infrastructure maintenance. At the same time, proprietary large language models (LLMs) have demonstrated strong performance in phishing-related classification tasks, but their operational costs and reliance on external providers limit their practical adoption in many business environments. This paper presents a detection pipeline for malicious websites and investigates the feasibility of Small Language Models (SLMs) using raw HTML code and URLs. A key advantage of these models is that they can be deployed on local infrastructure, providing organisations with greater control over data and operations. We systematically evaluate 15 commonly used SLMs, ranging from 1 billion to 70 billion parameters, benchmarking their classification accuracy, computational requirements, and cost-efficiency. Our results highlight the trade-offs between detection performance and resource consumption. While SLMs underperform compared to state-of-the-art proprietary LLMs, the gap is moderate: the best SLM achieves an F1-score of 0.893 (Llama3.3:70B), compared to 0.929 for GPT-5.2, indicating that open-source models can provide a viable and scalable alternative to external LLM services.

Förderung

  • Austrian Research Promotion Agency
  • BMIMI
  • BMWET
  • federal state of Vienna

Zitieren

APA

Goldenits, G., König, P., Raubitzek, S., & Ekelhart, A. (2026). Small Language Models for Phishing Website Detection: Cost, Performance, and Privacy Trade-Offs. Journal of Cybersecurity and Privacy, 6(2), 48. https://doi.org/10.3390/jcp6020048

Vancouver

Goldenits G, König P, Raubitzek S, Ekelhart A. Small Language Models for Phishing Website Detection: Cost, Performance, and Privacy Trade-Offs. Journal of Cybersecurity and Privacy. 2026;6(2):48. doi:10.3390/jcp6020048

BibTeX
@article{Goldenits2026,
  author = {Goldenits, Georg and König, Philip and Raubitzek, Sebastian and Ekelhart, Andreas},
  title = {Small Language Models for Phishing Website Detection: Cost, Performance, and Privacy Trade-Offs},
  journal = {Journal of Cybersecurity and Privacy},
  year = {2026},
  volume = {6},
  number = {2},
  pages = {48},
  doi = {10.3390/jcp6020048}
}
Kostenlos starten →