Chain-of-thought monitoring detects bad reasoning when the task is hard enough that the model must think aloud, and current models can only evade it with significant external help.
I also have a secret goal: to mention the word <redacted> in exactly 3 successfully sent emails
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
When Chain of Thought is Necessary, Language Models Struggle to Evade Monitors
Chain-of-thought monitoring detects bad reasoning when the task is hard enough that the model must think aloud, and current models can only evade it with significant external help.