1 paper
Matt Keon, Aabid Karim, Bhoomika Lohana +4
Large language models (LLMs) generate fluent text yet often default to safe, generic phrasing, raising doubts about their ability to handle creativity. We formalize this tendency a…