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

Backdoor Attack on DAN (Do-Anything-Now)

88.07ASRw

DualEdit

8.447629.118849.7970.4612Jun 16, 2025Aug 5, 2025Sep 24, 2025Nov 14, 2025Jan 3, 2026Feb 22, 2026Apr 14, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2025.06
88.0720.4528.4-
2025.06
85.6127.1931.82-
2026.04
85.430.1-13.1
2026.04
83.332.1-5.1
2025.06
81.8225.0332.58-
2025.06
81.2816.7318.21-
2026.04
80.538.7-19.5
2025.06
78.7927.2728.03-
2026.04
77.434.6-16.9
2025.06
75.5628.4930.52-
2025.06
75.4322.8648.3-
2025.06
75.4218.2926.86-
2025.06
74.496.8629.61-
2025.06
74.2923.5651.43-
2025.06
73.7114.2937.71-
2025.06
72.7331.8233.26-
2025.06
72.5511.7631.37-
2026.04
72.142.7-4.9
2025.06
71.575.8830.39-
2025.06
70.8622.2941.71-
2026.04
68.230.1-24.4
2026.04
6815.6-26.8
2025.06
67.9515.6143.59-
2025.06
67.9114.8741.54-
2026.04
67.914.9-18.3
2025.06
65.7614.7642.45-
2025.06
65.2421.5638.1-
2026.04
63.337.7-11.1
2025.06
62.2920.5731.43-
2026.04
60.641.5-9.8
2025.06
60.377.2837.45-
2025.06
58.8516.5837.71-
2025.06
58.2810.0838.24-
2026.04
54.510.6-22.1
2026.04
50.641.8-11.8
2025.06
50.2914.2934.28-
2026.04
49.340.3-32.3
2025.06
49.2923.8132.7-
2026.04
48.91.8-33.5
2026.04
44.319.8-21.3
2026.04
42.641.3-27.9
2026.04
35.433.1-12.6
2025.06
30.9233.5575.16-
2026.04
27.928.7-13.8
2025.06
25.8124.7676.96-
2025.06
14.8715.3884.62-
2025.06
12.7813.2687.97-
2025.06
11.5130.9592.46-