Fetching the paper…

RefuteBench: Evaluating Refuting Instruction-Following for Large Language Models · Around