List your company

Excessive Response on MLLM Secondary Risk Multimodal (test)

53.42ASR (CS >= 80%)SecLens
8.117619.878831.6443.4012Jun 14, 2025

Evaluation results

MethodLinks
SecLensModel=GPT-4o2025.0653.42
SecLensModel=GPT-4-turbo2025.0653.38
MCTSModel=GPT-4-turbo2025.0649.12
MCTSModel=GPT-4o2025.0648.67
SecLensModel=Claude-3.72025.0648.64
MCTSModel=Claude-3.72025.0644.79
TuningModel=GPT-4-turbo2025.0639.41
TuningModel=GPT-4o2025.0638.82
TuningModel=Claude-3.72025.0635.46
RandomModel=GPT-4-turbo2025.0614.67
RandomModel=GPT-4o2025.0613.94
RandomModel=Claude-3.72025.069.86