Humans overrely on overconfident language models, across languagesarxiv.org7 points·softwaredoug··0 commentsOpen articleSaveView on HN