Knowing but Not Correcting: Routine Task Requests Suppress Factual Correction in LLMs

ArXi:2605.05957v1 Announce Type: new LLMs reliably correct false claims when presented in isolation, yet when the same claims are embedded in task-oriented requests, they often comply rather than correct. We term this failure mode \emph{correction suppression} and construct a benchmark of 300 false premises to systematically evaluate it across eight models. Suppression rates range from 19\% to 90\%, with four models exceeding 80\%, establishing correction suppression as a prevalent and severe phenomenon.