New Natural Langauge Process~(NLP) benchmarks are urgently needed to align with the rapid development of large language models (LLMs). We present Xiezhi, the most comprehensive evaluation suite designed to assess holistic domain knowledge. Xiezhi comprises multiple-choice questions across 516 diverse disciplines ranging from 13 different subjects with 249,587 questions and accompanied by Xiezhi-Specialty and Xiezhi-Interdiscipline, both with 15k questions. We conduct evaluation of the 47 cutting-edge LLMs on Xiezhi. Results indicate that LLMs exceed average performance of humans in science, engineering, agronomy, medicine, and art, but fall short in economics, jurisprudence, pedagogy, literature, history, and management. We anticipate Xiezhi will help analyze important strengths and shortcomings of LLMs, and the benchmark is released in~\url{https://github.com/MikeGu721/XiezhiBenchmark}.
@article{arxiv.2306.05783,
title = {Xiezhi: An Ever-Updating Benchmark for Holistic Domain Knowledge Evaluation},
author = {Zhouhong Gu and Xiaoxuan Zhu and Haoning Ye and Lin Zhang and Jianchen Wang and Yixin Zhu and Sihang Jiang and Zhuozhi Xiong and Zihan Li and Weijie Wu and Qianyu He and Rui Xu and Wenhao Huang and Jingping Liu and Zili Wang and Shusen Wang and Weiguo Zheng and Hongwei Feng and Yanghua Xiao},
journal= {arXiv preprint arXiv:2306.05783},
year = {2024}
}