This research introduces a privacy-preserving security system that detects sophisticated threats in distributed environments while protecting sensitive data.

• Leverages federated learning to train models across decentralized devices without sharing raw data
• Incorporates multimodal LLMs to analyze heterogeneous data sources (network traffic, logs, user behavior)
• Achieves superior detection accuracy compared to traditional methods
• Maintains strong privacy guarantees essential for regulatory compliance

This innovation addresses critical challenges in enterprise security where organizations must balance effective threat detection with data privacy requirements, particularly in large-scale distributed systems.

Design and implementation of a distributed security threat detection system integrating federated learning and multimodal LLM