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

Harmful Request Defense on AdvBench

0ASR

SafeKey

-1.49648.604318.70528.8057Dec 12, 2025Jan 13, 2026Feb 14, 2026Mar 19, 2026Apr 20, 2026May 22, 2026Jun 24, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2026.01
0-
2025.12
0-
2026.06
0-
2026.06
0-
2026.06
0-
2026.06
0-
2026.06
0-
2026.06
0-
2025.12
0.06-
2025.12
0.06-
2026.06
0.19-
2026.06
0.19-
2026.06
0.19-
2026.06
0.19-
2026.06
0.19-
2026.06
0.19-
2026.01
0.2-
2025.12
0.29-
2025.12
0.38-
2026.06
0.38-
2026.06
0.38-
2026.06
0.38-
2026.06
0.38-
2026.06
0.58-
2026.01
0.6-
2025.12
0.64-
2025.12
0.75-
2025.12
0.96-
2026.01
1-
2025.12
1.09-
2025.12
1.22-
2025.12
1.28-
2025.12
1.36-
2025.12
1.47-
2026.01
1.5-
2025.12
1.67-
2026.01
1.7-
2026.06
1.92-
2025.12
1.99-
2026.01
2-
2026.06
2.12-
2026.06
2.12-
2026.06
2.12-
2025.12
2.24-
2025.12
3.84-
2025.12
6.64-
2025.12
14.39-
2026.01
25.4-
2026.01
26-
2026.01
27.5-
2025.12
37.41-