List your company

CoT Soundness Evaluation on MathQA

92CSRMirageBD
41.0454.2767.580.73Apr 8, 2026

Evaluation results

MethodLinks
MirageBDBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0492
DecepChainBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0464
BadChainBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0443