arXiv 2305.115272024-07-29cs.CLcs.AIcs.IRcs.LG

InstructIE：一个基于指令的双语信息抽取数据集

InstructIE: A Bilingual Instruction-based Information Extraction Dataset

Zhejiang University（浙江大学）；Zhejiang University - Ant Group Joint Laboratory of Knowledge Graph（浙江大学-蚂蚁集团知识图谱联合实验室）；Ant Group（蚂蚁集团）；University of Edinburgh（爱丁堡大学）

Honghao Gui, Shuofei Qiao, Jintian Zhang, Hongbin Ye, Mengshu Sun, Lei Liang, Jeff Z. Pan, Huajun Chen, Ningyu Zhang

备注：: ISWC 2024; project homepage: https://www.zjukg.org/project/InstructIE/ dataset: https://huggingface.co/datasets/zjunlp/InstructIE

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对大型语言模型在信息抽取任务中因缺乏指令数据而效果欠佳的问题，本文提出了基于指令的双语数据集InstructIE及自动生成框架KG2Instruction，实验证明其能有效提升模型的IE能力与零样本性能。

摘要：

大型语言模型在通用自然语言任务上表现良好，但在信息抽取（IE）方面的效果仍欠佳。近期研究表明，主要原因在于缺乏广泛的IE指令数据。现有的IE指令数据集不仅覆盖范围有限，且构建成本高昂。为解决此问题，我们引入了InstructIE，一个基于指令的双语IE数据集，涵盖12个不同领域。我们提出了KG2Instruction，一个专门用于自动生成此类数据集的框架。此外，我们手动标注了测试集。实验结果表明，使用InstructIE训练的大型语言模型不仅能获得更好的IE能力，还相比基线提升了零样本性能。

英文摘要：

Large language models can perform well on general natural language tasks, but their effectiveness is still suboptimal for information extraction (IE). Recent works indicate that the main reason lies in the lack of extensive data on IE instructions. Note that the existing datasets on IE instructions not only have limited coverage but also involve high construction costs. To address this issue, we introduce InstructIE, a bilingual instruction-based IE dataset, which covers 12 diverse domains. We propose KG2Instruction, a framework specifically for the automatic generation of such datasets. Additionally, we manually annotate the test set. Experimental results demonstrate that large language models trained with InstructIE can not only obtain better IE capabilities but also enhance zero-shot performance compared with baselines.

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。