Authors: Pedro Miguel Sánchez Sánchez, Alberto Huertas Celdrán, Gérôme Bovet, Gregorio Martínez Pérez

Published on: May 15, 2024

Impact Score: 7.4

Arxiv code: Arxiv:2405.09318

Summary

• What is new: Integrates large language models with system call analysis to enhance malware detection.
• Why this is important: Traditional ML/DL methods for malware detection struggle with context and intent of attacks.
• What the research proposes: A novel framework using transfer learning on large language models to improve malware detection accuracy.
• Results: Achieved superior accuracy and F1-Score of approximately 0.86 using models like BigBird and Longformer.

Technical Details

Technological frameworks used: Transfer learning

Models used: BigBird, Longformer

Data used: Over 1TB of benign and malicious system call data

Potential Impact

Cybersecurity providers, military defense contractors, and secure communication services

Want to implement this idea in a business?

We have generated a startup concept here: CyberGuardAI.