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

TL;DR

This paper introduces YAYI-UIE, a chat-enhanced instruction tuning framework that improves universal information extraction for both Chinese and English, achieving state-of-the-art results especially in Chinese datasets.

Contribution

The paper presents a novel end-to-end framework that leverages dialogue and extraction data to enhance multilingual information extraction capabilities.

Findings

Achieves state-of-the-art performance on Chinese datasets.

Performs comparably on English datasets in supervised and zero-shot settings.

Effectively handles heterogeneous data structures and task-specific schemas.

Abstract

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.