arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.13889cs.CL
收藏

结合关系分类器的大语言模型用于文档级关系抽取

LLM with Relation Classifier for Document-Level Relation Extraction

School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳)计算机科学与技术学院);School of Computer Science and Electronic Engineering, University of Essex(埃塞克斯大学计算机科学与电子工程学院)

Xingzuo Li, Kehai Chen, Yunfei Long, Min Zhang

AI总结:

针对大语言模型在文档级关系抽取任务上性能落后的问题,提出先用分类器筛选潜在关系实体对再输入LLM分类的方法,显著提升了性能并缩小与最优BERT类模型的差距。

摘要:

大语言模型(LLMs)为自然语言处理开创了新范式。尽管LLMs取得了诸多进展,但在文档级关系抽取(DocRE)这一理解长上下文中复杂实体关系的关键任务上,基于LLM的方法仍落后于传统方法。本文探究了这一性能差距的成因,指出无关系实体对导致LLM注意力分散是关键因素。随后我们提出了一种新颖的分类器-LLM文档级关系抽取方法。具体而言,该方法首先通过一个分类器筛选出具有潜在关系的实体对候选,再将其输入LLM进行最终的关系分类。该方法可确保LLM在推理时将注意力集中在表达关系的实体对上,而非无关系的实体对。在DocRE基准上的实验表明,我们的方法显著优于近期基于LLM的DocRE模型,并缩小了与基于BERT的最优模型之间的性能差距。

英文摘要:

Large language models (LLMs) have created a new paradigm for natural language processing. Despite their advancement, LLM-based methods still lag behind traditional approaches in document-level relation extraction (DocRE), a critical task for understanding complex entity relations within long context. This paper investigates the causes of this performance gap, identifying the dispersion of attention by LLMs due to entity pairs without relations as a key factor. We then introduce a novel classifier-LLM approach to DocRE. Particularly, the proposed approach begins with a classifier designed to select entity pair candidates that exhibit potential relations and then feed them to LLM for final relation classification. This method ensures that the LLM's attention is directed at relation-expressing entity pairs instead of those without relations during inference. Experiments on DocRE benchmarks reveal that our method significantly outperforms recent LLM-based DocRE models and narrows the performance gap with state-of-the-art BERT-based models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑