autonomous-defense

#autonomous-defense

Distilling Knowledge from Large Language Models into Lightweight Reinforcement Learning Agents for Autonomous Cyber Operations

arXiv cs.LG ↗ · 2026-08-03 Cached

This paper investigates using an 8-billion parameter LLM to improve autonomous cyber defense, then distills its policy into a lightweight 64,910-parameter RL agent, demonstrating feasibility across CybORG scenarios.

0 favorites 0 likes

autonomous-defense

Distilling Knowledge from Large Language Models into Lightweight Reinforcement Learning Agents for Autonomous Cyber Operations

Submit Feedback