CKnowEdit: A New Chinese Knowledge Editing Dataset for Linguistics, Facts, and Logic Error Correction in LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fang, Jizhan, Lu, Tianhe, Yao, Yunzhi, Jiang, Ziyan, Xu, Xin, Chen, Huajun, Zhang, Ningyu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916770761146368
author Fang, Jizhan
Lu, Tianhe
Yao, Yunzhi
Jiang, Ziyan
Xu, Xin
Chen, Huajun
Zhang, Ningyu
author_facet Fang, Jizhan
Lu, Tianhe
Yao, Yunzhi
Jiang, Ziyan
Xu, Xin
Chen, Huajun
Zhang, Ningyu
contents Chinese, as a linguistic system rich in depth and complexity, is characterized by distinctive elements such as ancient poetry, proverbs, idioms, and other cultural constructs. However, current Large Language Models (LLMs) face limitations in these specialized domains, highlighting the need for the development of comprehensive datasets that can assess, continuously update, and progressively improve these culturally-grounded linguistic competencies through targeted training optimizations. To address this gap, we introduce CKnowEdit, the first-ever Chinese knowledge editing dataset designed to correct linguistic, factual, and logical errors in LLMs. We collect seven types of knowledge from a wide range of sources, including classical texts, idioms, and content from Baidu Tieba Ruozhiba, taking into account the unique polyphony, antithesis, and logical structures inherent in the Chinese language. By analyzing this dataset, we highlight the challenges current LLMs face in mastering Chinese. Furthermore, our evaluation of state-of-the-art knowledge editing techniques reveals opportunities to advance the correction of Chinese knowledge. Code and dataset are available at https://github.com/zjunlp/EasyEdit.
format Preprint
id arxiv_https___arxiv_org_abs_2409_05806
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CKnowEdit: A New Chinese Knowledge Editing Dataset for Linguistics, Facts, and Logic Error Correction in LLMs
Fang, Jizhan
Lu, Tianhe
Yao, Yunzhi
Jiang, Ziyan
Xu, Xin
Chen, Huajun
Zhang, Ningyu
Computation and Language
Artificial Intelligence
Information Retrieval
Machine Learning
Chinese, as a linguistic system rich in depth and complexity, is characterized by distinctive elements such as ancient poetry, proverbs, idioms, and other cultural constructs. However, current Large Language Models (LLMs) face limitations in these specialized domains, highlighting the need for the development of comprehensive datasets that can assess, continuously update, and progressively improve these culturally-grounded linguistic competencies through targeted training optimizations. To address this gap, we introduce CKnowEdit, the first-ever Chinese knowledge editing dataset designed to correct linguistic, factual, and logical errors in LLMs. We collect seven types of knowledge from a wide range of sources, including classical texts, idioms, and content from Baidu Tieba Ruozhiba, taking into account the unique polyphony, antithesis, and logical structures inherent in the Chinese language. By analyzing this dataset, we highlight the challenges current LLMs face in mastering Chinese. Furthermore, our evaluation of state-of-the-art knowledge editing techniques reveals opportunities to advance the correction of Chinese knowledge. Code and dataset are available at https://github.com/zjunlp/EasyEdit.
title CKnowEdit: A New Chinese Knowledge Editing Dataset for Linguistics, Facts, and Logic Error Correction in LLMs
topic Computation and Language
Artificial Intelligence
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2409.05806