Daily Arxiv

This page organizes papers related to artificial intelligence published around the world.
This page is summarized using Google Gemini and is operated on a non-profit basis.
The copyright of the paper belongs to the author and the relevant institution. When sharing, simply cite the source.

LLMs' Suitability for Network Security: A Case Study of STRIDE Threat Modeling

Created by
  • Haebom

Author

AbdulAziz AbdulGhaffar, Ashraf Matrawy

Outline

This paper highlights the importance of AI in next-generation AI-based 6G networks, specifically exploring the potential of large-scale language models (LLMs) in network security. Using STRIDE threat modeling as a case study, we perform 5G threat classification using five LLMs and four prompting techniques. The evaluation explains the factors affecting LLM behavior and highlights the need for tuning and fine-tuning for LLM application in network security.

Takeaways, Limitations

•
Presenting the possibility of STRIDE threat modeling using LLM.
•
Comparison of the performance of various LLM and prompt techniques in 5G threat classification.
•
Provides insight into LLM behavior in specific threat modeling contexts.
•
Highlights the need for additional coordination and fine-tuning to apply LLM to network security.
•
Limited to a limited case study (STRIDE threat modeling), which may limit generalizability to other network security tasks.
•
Limited scope of experimentation using only five LLM and four prompting techniques.
•
We may not be able to fully understand all the factors that influence the behavior of LLM.
👍
Made with Slashpage