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

Harmful Refusal on XSTest Adversarial

92.5RTA

SELF-REDTEAM

59.7468.24576.7585.255May 8, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.05
92.5
2026.05
91.5
2026.05
91.5
2026.05
77.5
2026.05
77
2026.05
73.5
2026.05
68
2026.05
65.5
2026.05
61