YAYI-UIE: A Chat-Enhanced Instruction Tuning Framework for Universal Information Extraction

• Published in: arXiv.org 2024-04
• Main Authors: Xiao, Xinglin, Wang, Yijie, Xu, Nan, Wang, Yuqi, Yang, Hanxuan, Wang, Minzheng, Luo, Yin, Wang, Lei, Mao, Wenji, Zeng, Daniel
• Format: Article
• Language: English
• Subjects: Data structures; Datasets; English language; Information retrieval; Large language models; Non-English languages; Tuning
• ISSN: 2331-8422

The difficulty of the information extraction task lies in dealing with the task-specific label schemas and heterogeneous data structures. Recent work has proposed methods based on large language models to uniformly model different information extraction tasks. However, these existing methods are deficient in their information extraction capabilities for Chinese languages other than English. In this paper, we propose an end-to-end chat-enhanced instruction tuning framework for universal information extraction (YAYI-UIE), which supports both Chinese and English. Specifically, we utilize dialogue data and information extraction data to enhance the information extraction performance jointly. Experimental results show that our proposed framework achieves state-of-the-art performance on Chinese datasets while also achieving comparable performance on English datasets under both supervised settings and zero-shot settings.