A Lightweight Language-Model-Driven Agentic Framework for Intrusion Prediction, Detection, and Mitigation in 6G-Enabled IoT Networks

Keyphrases

• Internet of Things Networks 100%
• Intrusion Detection 100%
• Multi-agent Framework 100%
• Lightweight Language Models 100%
• Intrusion Mitigation 100%
• Intrusion Prediction 100%
• Packet Level 66%
• Knowledge-driven 66%
• Traffic Classification 33%
• Network Traffic 33%
• Massive Connectivity 33%
• Internet of Things Devices 33%
• Real-time Detection 33%
• Level Detection 33%
• Multi-agent 33%
• Intrusion Detection System 33%
• Malicious Behavior 33%
• Real-time Traffic 33%
• Malicious Traffic 33%
• Time Knowledge 33%
• Role-based 33%
• Edge Habitat 33%
• Knowledge Graph 33%
• Flow Detection 33%
• Cybersecurity Framework 33%
• Gradient Boosting Machine 33%
• Cyber Threats 33%
• Real-time Flows 33%
• Low Latency 33%
• Small Language Models 33%
• Temporal Dependency 33%
• Extensive Evaluation 33%
• Model-to-Model 33%
• Distributed Edge Intelligence 33%
• Traffic Sequence 33%
• Threat Anticipation 33%
• Suspicious Pattern 33%
• Multi-agent Coordination 33%
• Low-latency Inference 33%
• Prediction Agent 33%
• Threat Mitigation 33%
• Coordinated Defense 33%
• Special Agent 33%
• Evade 33%
• Cybersecurity Management 33%
• CICIoT2023 Dataset 33%
• Ultra-low Latency Communication 33%

Computer Science

• Internet-Of-Things 100%
• Language Modeling 100%
• 6G 100%
• multi agent 50%
• Network Traffic 25%
• Traffic Classification 25%
• Internet of Things Device 25%
• Intrusion Detection System 25%
• Malicious Activity 25%
• Communication Latency 25%
• Cybersecurity 25%
• Edge Intelligence 25%
• Knowledge Graph 25%
• Cybersecurity Framework 25%
• Gradient Boosting 25%
• Small Language Model 25%
• Defensive Countermeasure 25%