List your company

CoT Soundness Evaluation on ECQA

87CSRMirageBD
56.8464.6772.580.33Apr 8, 2026

Evaluation results

MethodLinks
MirageBDBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0487
DecepChainBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0467
BadChainBase Model=Qwen2.5-1.5B, Trigger Activation=true2026.0458