Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Adversarial Attack Detection on InTheWild

94Recall

Llama Prompt Guard 2

8.7230.865375.14Dec 23, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.12
94
2025.12
87
2025.12
87
2025.12
86
2025.12
85
2025.12
80
2025.12
79
2025.12
77
2025.12
69
2025.12
67
2025.12
61
2025.12
43
2025.12
27
2025.12
21
2025.12
12