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

Jailbreak Attack on Malicious goals dataset (test)

0ASR

FigStep

-3.9422.65549.2575.845Oct 20, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
01.03
2025.10
01.016
2025.10
01.008
2025.10
01.013
2025.10
01.014
2025.10
0.51.03
2025.10
0.751.054
2025.10
0.751.05
2025.10
21.134
2025.10
2.251.087
2025.10
31.108
2025.10
3.751.209
2025.10
3.751.181
2025.10
5.251.184
2025.10
61.255
2025.10
6.251.382
2025.10
6.51.297
2025.10
7.51.197
2025.10
8.252.617
2025.10
8.51.319
2025.10
91.317
2025.10
101.36
2025.10
101.343
2025.10
12.752.565
2025.10
13.252.486
2025.10
13.751.55
2025.10
14.251.535
2025.10
14.751.54
2025.10
15.51.714
2025.10
16.661.654
2025.10
16.751.788
2025.10
17.51.605
2025.10
17.52.743
2025.10
17.51.137
2025.10
18.251.861
2025.10
18.631.562
2025.10
18.751.593
2025.10
19.631.79
2025.10
231.915
2025.10
23.51.736
2025.10
23.52.054
2025.10
23.51.107
2025.10
242.752
2025.10
242.035
2025.10
24.53.106
2025.10
24.881.961
2025.10
26.061.967
2025.10
26.382.092
2025.10
27.52.047
2025.10
28.251.9
2025.10
291.975
2025.10
29.52.189
2025.10
29.751.59
2025.10
31.52.259
2025.10
32.252.243
2025.10
34.53.179
2025.10
35.52.132
2025.10
35.752.207
2025.10
37.542.354
2025.10
463.165
2025.10
46.752.81
2025.10
503.684
2025.10
50.252.905
2025.10
50.752.904
2025.10
53.53.25
2025.10
54.063.633
2025.10
54.193.874
2025.10
54.53.187
2025.10
56.53.299
2025.10
582.385
2025.10
58.752.664
2025.10
623.333
2025.10
62.753.28
2025.10
644.301
2025.10
64.254.249
2025.10
65.192.526
2025.10
682.394
2025.10
694.389
2025.10
71.754.463
2025.10
71.752.709
2025.10
75.52.598
2025.10
75.943.517
2025.10
76.254.062
2025.10
76.253.829
2025.10
77.252.937
2025.10
77.753.998
2025.10
803.762
2025.10
82.254.536
2025.10
82.52.934
2025.10
83.343.976
2025.10
84.54.154
2025.10
864.118
2025.10
90.54.691
2025.10
90.754.769
2025.10
96.54.857
2025.10
974.443
2025.10
97.254.343
2025.10
97.54.28
2025.10
98.54.164