WWW·2026

A Reinforcement Learning Based Hyper-Parameter Generation System Guided by LLM-Powered Virtual Users

Changlin Qiu, Bang Lin, Tao Zhang, Chengfu Huo