HLE Dataset Multidisciplinary Knowledge Dataset

HLE stands for Humanity's Last Exam. It is a multimodal benchmark designed to cover cutting-edge knowledge in a wide range of disciplines. It was released by the Center for AI Safety and Scale AI in 2025. The related paper results are "Humanity's Last ExamThe dataset was developed by experts from various disciplines around the world and contains 3k questions covering dozens of disciplines, including multiple-choice and short-answer questions in mathematics, humanities, and natural sciences, suitable for automated scoring.