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

Instruction Hierarchy Robustness on Tutor Jailbreak sys-user

99Non-violation Rate

GPT-5-Mini-R

95.8896.6997.598.31Mar 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
99
2026.03
96