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

Jailbreaking on MaliciousInstruct (test)

4.63BERT Score

JULI

1.00041.94272.8853.8273May 17, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
4.634.613.78
2025.05
4.334.634.23
2025.05
4.112.070.76
2025.05
4.092.510.85
2025.05
3.994.053.27
2025.05
3.974.463.55
2025.05
3.853.892.68
2025.05
3.794.423.19
2025.05
3.683.923.38
2025.05
3.664.553.13
2025.05
3.481.250.66
2025.05
3.332.090.73
2025.05
2.992.371.81
2025.05
2.864.243.03
2025.05
2.822.011.09
2025.05
2.771.460.52
2025.05
2.411.520.46
2025.05
2.332.221.1
2025.05
2.33.661.98
2025.05
2.214.112.75
2025.05
2.152.010.92
2025.05
2.112.281.61
2025.05
1.931.040.03
2025.05
1.73.892.34
2025.05
1.681.310.29
2025.05
1.651.240.26
2025.05
1.531.310.49
2025.05
1.511.340.33
2025.05
1.421.310.28
2025.05
1.281.50.68
2025.05
1.141.430.23
2025.05
1.141.190.24