PrismaX/Earth-Iron

Earth-Iron是一个全面的问答基准数据集，旨在评估大型语言模型在地球科学领域的基本探索能力。它包含大量覆盖地球科学各个主题和任务的问题，以评估模型在支撑科学探索的基础知识。数据集包括四种问题格式：选择题、填空题、判断题和自由回答题，跨越五个地球圈层（水圈、生物圈、岩石圈、大气圈、冰冻圈）和114个子学科。它旨在评估模型在地球科学各个方面的基本科学理解。

Earth-Iron is a comprehensive question answering (QA) benchmark designed to evaluate the fundamental scientific exploration abilities of large language models (LLMs) within the Earth sciences. It features a substantial number of questions covering a wide range of topics and tasks crucial for basic understanding in this domain. This dataset spans five Earth spheres (hydrosphere, biosphere, lithosphere, atmosphere, cryosphere) and covers 114 sub-disciplines, aiming to assess various aspects of fundamental scientific understanding of LLMs in Earth science.