“Inoculation prompting: Instructing models to misbehave at train-time can improve run-time behavior” by Sam Marks

by LessWrong (Curated & Popular)