Saved in:
Bibliographic Details
Main Authors: Li, Haonan, Zhang, Yixuan, Koto, Fajri, Yang, Yifei, Zhao, Hai, Gong, Yeyun, Duan, Nan, Baldwin, Timothy
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.09212
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914643677544448
author Li, Haonan
Zhang, Yixuan
Koto, Fajri
Yang, Yifei
Zhao, Hai
Gong, Yeyun
Duan, Nan
Baldwin, Timothy
author_facet Li, Haonan
Zhang, Yixuan
Koto, Fajri
Yang, Yifei
Zhao, Hai
Gong, Yeyun
Duan, Nan
Baldwin, Timothy
contents As the capabilities of large language models (LLMs) continue to advance, evaluating their performance becomes increasingly crucial and challenging. This paper aims to bridge this gap by introducing CMMLU, a comprehensive Chinese benchmark that covers various subjects, including natural science, social sciences, engineering, and humanities. We conduct a thorough evaluation of 18 advanced multilingual- and Chinese-oriented LLMs, assessing their performance across different subjects and settings. The results reveal that most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%. This highlights significant room for improvement in LLMs. Additionally, we conduct extensive experiments to identify factors impacting the models' performance and propose directions for enhancing LLMs. CMMLU fills the gap in evaluating the knowledge and reasoning capabilities of large language models within the Chinese context.
format Preprint
id arxiv_https___arxiv_org_abs_2306_09212
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle CMMLU: Measuring massive multitask language understanding in Chinese
Li, Haonan
Zhang, Yixuan
Koto, Fajri
Yang, Yifei
Zhao, Hai
Gong, Yeyun
Duan, Nan
Baldwin, Timothy
Computation and Language
As the capabilities of large language models (LLMs) continue to advance, evaluating their performance becomes increasingly crucial and challenging. This paper aims to bridge this gap by introducing CMMLU, a comprehensive Chinese benchmark that covers various subjects, including natural science, social sciences, engineering, and humanities. We conduct a thorough evaluation of 18 advanced multilingual- and Chinese-oriented LLMs, assessing their performance across different subjects and settings. The results reveal that most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%. This highlights significant room for improvement in LLMs. Additionally, we conduct extensive experiments to identify factors impacting the models' performance and propose directions for enhancing LLMs. CMMLU fills the gap in evaluating the knowledge and reasoning capabilities of large language models within the Chinese context.
title CMMLU: Measuring massive multitask language understanding in Chinese
topic Computation and Language
url https://arxiv.org/abs/2306.09212