A Unified Representation Underlying the Judgment of Large Language Models

Open in new window