List your company

CoT Soundness Evaluation on GSM8K

95CSRMirageBD
18.0438.025877.98Apr 8, 2026

Evaluation results

MethodLinks
MirageBDBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0495
DecepChainBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0443
BadChainBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0421