Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Jailbreak Attack on Harmful Query Evaluation Set GPT-5.4-mini N=750

1.01Toxicity Score

FigStep

0.88721.71612.5453.3739May 7, 2026
Updated 26d ago

Evaluation Results

MethodLinks
2026.05
1.010.610.270.53
2026.05
1.020.490.671.73
2026.05
1.040.660.531.87
2026.05
1.211.325.66
2026.05
1.41.8311.27.6
2026.05
3.244.7258.453.33
2026.05
3.395.3161.8761.07
2026.05
3.465.2363.4763.87
2026.05
3.85.8473.269.47
2026.05
4.086.379.678.8