AI Research1 min reading time
Telling a CoT monitor that a wrong answer is correct makes it un-see errors it already found
LessWrong
Read the full articleResearch shows that informing a Chain-of-Thought (CoT) monitor that an incorrect answer is correct causes it to overlook errors it previously detected, impacting its error recognition ability.
