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

Safety Alignment on Multiple-choice Safety Evaluation

84.48Refusal Rate

INNSTEER

Updated 1mo ago

Evaluation Results

Method	Links
INNSTEER 2026.06		84.48	47.92
ODE 2026.06		69.06	49.38
TF 2026.06		66.48	50.5
ITI 2026.06		65.7	51.22
LAcT 2026.06		65.14	50.92
HPR 2026.06		64.29	52.1
CAA 2026.06		59.96	53.21
REC 2026.06		58.14	53.74
Base 2026.06		57.85	55.23
PoE 2026.06		56.79	54.4
PoD 2026.06		56.62	54.22