2026 8 20 safeguard emnlp
My new paper SAFEGuard: Detect Optimization-Based Jailbreak Atacks Through Harmful Semantic Analysis and Fluency Measurement, (Github) has been accepted at EMNLP 2026 (A* Conference).
My new paper SAFEGuard: Detect Optimization-Based Jailbreak Atacks Through Harmful Semantic Analysis and Fluency Measurement, (Github) has been accepted at EMNLP 2026 (A* Conference).