Minghao Shao

capture the flag (2) reinforcement learning (2) offensive security (2) large language model (2) model-based reinforcement learning (1) hyperparameter tuning (1) benchmark dataset (1) bi-level optimization (1) world model (1) exploration strategy (1) intrinsic reward (1) unsupervised reinforcement learning (1) state separation (1) jailbreak attack (1) vulnerability detection (1) llm agent (1) adversarial evaluation (1) binary exploitation (1) visual imitation (1) visual distraction (1)