{"title":"Zero-shot relation triplet extraction as Next-Sentence Prediction","authors":"","doi":"10.1016/j.knosys.2024.112507","DOIUrl":null,"url":null,"abstract":"<div><div>Zero-shot relation triplet extraction (ZeroRTE) endeavors to extract relation triplets from a test set using a model trained on a training set with disjoint relations from the test set. Current ZeroRTE approaches primarily rely on two strategies: 1) Combining pre-trained language models to generate additional training samples; 2) Adding a large number of parameters that require training from scratch on top of a pre-trained language model. However, the former approach does not ensure the quality of generated samples, and the latter often struggles to generalize to unseen relations in the test set, particularly when the training set is small. In this paper, we introduce a novel method, <u>N</u>ext <u>S</u>entence <u>P</u>rediction for <u>R</u>elation <u>T</u>riplet <u>E</u>xtraction (NSP-RTE), abstracting ZeroRTE as a higher-level next sentence prediction (NSP) task to enhance its generalization ability to unseen relation categories. NSP-RTE integrates modules for relation recognition, entity detection, and triplet classification, leveraging pre-trained BERT models with fewer parameters requiring training from scratch, while eliminating the need for additional sample generation. Our experiments on the FewRel and Wiki-ZSL datasets demonstrate that NSP-RTE, with its simple and efficient design, significantly outperforms previous methods.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":null,"pages":null},"PeriodicalIF":7.2000,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Knowledge-Based Systems","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0950705124011419","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
Zero-shot relation triplet extraction (ZeroRTE) endeavors to extract relation triplets from a test set using a model trained on a training set with disjoint relations from the test set. Current ZeroRTE approaches primarily rely on two strategies: 1) Combining pre-trained language models to generate additional training samples; 2) Adding a large number of parameters that require training from scratch on top of a pre-trained language model. However, the former approach does not ensure the quality of generated samples, and the latter often struggles to generalize to unseen relations in the test set, particularly when the training set is small. In this paper, we introduce a novel method, Next Sentence Prediction for Relation Triplet Extraction (NSP-RTE), abstracting ZeroRTE as a higher-level next sentence prediction (NSP) task to enhance its generalization ability to unseen relation categories. NSP-RTE integrates modules for relation recognition, entity detection, and triplet classification, leveraging pre-trained BERT models with fewer parameters requiring training from scratch, while eliminating the need for additional sample generation. Our experiments on the FewRel and Wiki-ZSL datasets demonstrate that NSP-RTE, with its simple and efficient design, significantly outperforms previous methods.
期刊介绍:
Knowledge-Based Systems, an international and interdisciplinary journal in artificial intelligence, publishes original, innovative, and creative research results in the field. It focuses on knowledge-based and other artificial intelligence techniques-based systems. The journal aims to support human prediction and decision-making through data science and computation techniques, provide a balanced coverage of theory and practical study, and encourage the development and implementation of knowledge-based intelligence models, methods, systems, and software tools. Applications in business, government, education, engineering, and healthcare are emphasized.