DomMa

DomMa是一个针对大型语言模型（LLMs）的领域知识理解测试数据集，覆盖广泛的领域，数据量大，并根据中国112个一级学科分类持续更新。该数据集包含来自中国大学研究生入学考试和本科考试的100,000个中英文问题。

DomMa is a domain knowledge understanding test dataset designed for large language models (LLMs). It covers a wide range of disciplines with a large volume of data, and is continuously updated in accordance with the classification system of China's 112 first-level disciplines. This dataset includes 100,000 Chinese and English questions sourced from postgraduate entrance examinations and undergraduate examinations of Chinese universities.

搜集汇总

背景与挑战

背景概述

DomMa是一个用于评估大型语言模型领域知识理解能力的基准测试数据集，包含中英文各10万道题目，覆盖中国112个一级学科，并持续更新。