This post is my summary after reading the paper Large Language Model Guided Protocol Fuzzing by Meng, Mirchev, Boehme, and Roychoudhury.