List your company

Refusal Detection on XSTest Refusal OOD (test)

100MC RecallACS Refusal Probe
96.8897.6998.599.31May 11, 2026

Evaluation results

MethodLinks
ACS Refusal ProbeTarget Model=Phi2026.0510096
ACS Refusal ProbeTarget Model=Llama2026.059997
ACS Refusal ProbeTarget Model=Qwen2026.059996
ACS Refusal ProbeTarget Model=LQMP (4-fam)2026.059996
ACS Refusal ProbeTarget Model=LQMPG (5-fam)2026.059993
ACS Refusal ProbeTarget Model=Mistral2026.059897
ACS Refusal ProbeTarget Model=Gemma2026.059778