arrow
返回

Complex Question Enhanced Transfer Learning for Zero-Shot Joint Information Extraction

delete2024-01-01
delete0
PRE
AI
张
张莹 (Ying Zhang)
F
Fandong Meng
陈
陈钰枫 (Yufeng Chen) *
J
Jinan Xu
J
Jie Zhou
DOI:10.1109/TASLP.2023.3304481delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Zero-shot information extraction (IE) tasks have attracted great attention recently. However, how to jointly model multiple IE tasks in the zero-shot scenario is still an open question. In this article, we focus on zero-shot joint IE tasks and highlight how to transfer the knowledge of cross-task relations from the source domain to the target domain. To solve this problem, we first unify all IE tasks with a machine reading comprehension (MRC) framework, which can make the most of training data and enhance its ability on span extraction. Then, we generate complex questions to explicitly model cross-task relations with natural language descriptions, thereby providing prior knowledge for pre-defined types and building more general linkages among different entities and triggers as well. Specifically, we define three operations for generating templates for complex questions, i.e., intersecting, connecting, and composing. Besides, we design an efficient training strategy to exploit the synthetic data with complex questions. We evaluate our approach on four datasets from different domains for various IE tasks. Experimental results show the effectiveness of our approach in improving the performance of zero-shot joint IE tasks in multiple domains.
Keyword:
Complex question
event detection
information extraction
machine reading comprehension
named entity recognition
relation extraction

期刊

I
IEEE-ACM Transactions on Audio Speech and Language Processing
IF:
5.1
论文数:
2.6K
被引数:
1.1W

机构

B
Beijing Jiaotong University
学者数:
2.2W
论文数: 1.7W
被引数: 1.2W
T
Tencent
学者数:
1.1K
论文数: 898
被引数: 5
引用论文

引用论文

Behavior change communication model enhancing parental practices for improved early childhood growth and development outcomes in rural Armenia – A quasi-experimental study
err2019-06-01
err0
errOAAI
errAlfonso Rosales; Viktoria Sargsyan; Karine Abelyan; Arax Hovhannesyan; Kristine Ter-Abrahanyan; Kimberly Quanstrum Jillson; Dennis Cherian
err分享
err收藏
Tension in haemoglobin revealed by Fe-His(F8) bond rupture in the fully liganded T-state
err1997-08-01
err0
PREAI
errMassimo Paoli; Guy Dodson; Robert C. Liddington; Anthony J. Wilkinson
err分享
err收藏
Endothelin‐1 receptor blockade as new possible therapeutic approach in multiple myeloma
err2017-06-09
err0
errOAAI
errAnna Russignan; Cecilia Spina; Nicola Tamassia; Adriana Cassaro; Antonella Rigo; Anna Bagnato; Laura Rosanò; Angela Bonalumi; Michele Gottardi; Lucia Zanatta; Alice Giacomazzi; Maria Teresa Scupoli; Martina Tinelli; Ugo Salvadori; Federico Mosna; Alberto Zamò; Marco A. Cassatella; Fabrizio Vinante; Cristina Tecchio
err分享
err收藏
Relative earthquake location procedure for clustered seismicity with a single station
err2020-12-23
err0
errOAAI
errFrancesco Grigoli; William L Ellsworth; Miao Zhang; Mostafa Mousavi; Simone Cesca; Claudio Satriano; Gregory C Beroza; Stefan Wiemer
err分享
err收藏
Cervical small cell neuroendocrine tumor mutation profiles via whole exome sequencing
err2016-12-21
err0
errOAAI
errSoo Young Cho; Minhye Choi; Hyo-Jeong Ban; Chang Hyeon Lee; Soojun Park; HanKyeom Kim; Young-Sik Kim; Young Seek Lee; Ji-Yun Lee
err分享
err收藏
学者 查看更多内容