logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
arrow
返回

Learning rules and aligning elements for document-level relation extraction

delete2025-12-08
delete0
PRE
AI
G
Ganlin Xu
J
Jianzhou Feng
Q
Qin Wang
DOI：10.1016/j.ipm.2025.104511delete
delete原文链接
delete分享
delete收藏
摘要

摘要

En 中文
文档级关系抽取（DocRE）旨在推断文档中实体对1之间的语义关系。基于生成的方法在DocRE中仅从纯文本学习表层文本模式，而非逻辑规则模式，同时生成不受控的输出。因此，本文提出了一种新颖的生成范式，即一种规则学习与元素对齐（RLEA）方法用于DocRE。我们使用两个T5模型（文本学习器和规则学习器）构建对称结构，其中文本学习器从符号三元组学习文本模式，规则学习器从链式逻辑规则学习规则模式。为更好地解决上述挑战，我们提出了三项关键技术：双向门控函数、规则正则器和元素对齐机制。实验结果表明，我们的方法在关系抽取和逻辑一致性上取得了最佳结果，其中RLEA在DWIE数据集上Ign F1、F1和Logic指标分别达到72.37、79.44和94.52，在DocRED数据集上Ign F1和F1指标分别达到61.94和63.96，在Re-DocRED数据集上Ign F1和F1指标分别达到76.81和77.06。此外，定量实验和定性分析展示了逻辑规则如何作用于DocRE的基于生成的黑盒模型2。

期刊

I
Information Processing and Management
IF:
6.9
论文数:
5.2K
被引数:
1.5W

机构

H
Hunan University
学者数:
3.6K
论文数: 1.1K
被引数: 0
Y
yanshan university
学者数:
1.7K
论文数: 445
被引数: 0
F
fudan university
学者数:
7.2K
论文数: 1.9K
被引数: 0
学者 查看更多机构
引用论文

引用论文

Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling
errProceedings of the AAAI Conference on Artificial Intelligence
IF0
err2021-05-18
err0
errOAAI
errWenxuan Zhou; Kevin Huang; Tengyu Ma; Jing Huang
err分享
err收藏
LLMs for knowledge graph construction and reasoning: recent capabilities and future opportunities
errWorld Wide Web
IF0
err2024-08-21
err0
PREAI
errYuqi Zhu; Xiaohan Wang; Jing Chen; Shuofei Qiao; Yixin Ou; Yunzhi Yao; Shumin Deng; Huajun Chen; Ningyu Zhang
err分享
err收藏
Rule mining over knowledge graphs via reinforcement learning
errKNOWLEDGE-BASED SYSTEMS
IF7.6
err2022-04-01
err11
errOAAI
errChen, Lihan; Jiang, Sihang; Liu, Jingping; Wang, Chao; Zhang, Sheng; Xie, Chenhao; Liang, Jiaqing; Xiao, Yanghua; Song, Rui
err分享
err收藏
Multi-granularity sequential neural network for document-level biomedical relation extraction
errINFORMATION PROCESSING & MANAGEMENT
IF6.9
err2021-11-01
err11
PREAI
errLiu, Xiaofeng; Tan, Kaiwen; Dong, Shoubin
err分享
err收藏
Collective prompt tuning with relation inference for document-level relation extraction
errINFORMATION PROCESSING & MANAGEMENT
IF6.9
err2023-09-01
err7
PREAI
errYuan, Changsen; Cao, Yixin; Huang, Heyan
err分享
err收藏
Document-Level Relation Extraction with Reconstruction
errProceedings of the AAAI Conference on Artificial Intelligence
IF0
err2021-05-18
err0
errOAAI
errWang Xu; Kehai Chen; Tiejun Zhao
err分享
err收藏
Revisiting Document-Level Relation Extraction with Context-Guided Link Prediction基于上下文引导链接预测重新审视文档级关系抽取
errProceedings of the AAAI Conference on Artificial Intelligence
IF0
err
err0
PREAI
errJain,Monika; Mutharaju,Raghava; Kavuluru,Ramakanth; Singh,Kuldeep
err分享
err收藏
An adaptive confidence-based data revision framework for Document-level Relation Extraction
errINFORMATION PROCESSING & MANAGEMENT
IF6.9
err2025-01-01
err0
PREAI
errJiang, Chao; Liao, Jinzhi; Zhao, Xiang; Zeng, Daojian; Dai, Jianhua
err分享
err收藏
Optimal Thresholding of Classifiers to Maximize F1 Measure
errLecture Notes in Computer Science
IF0
err2014-01-01
err0
errOAAI
errZachary C. Lipton; Charles Elkan; Balakrishnan Naryanaswamy
err分享
err收藏
Document-level denoising relation extraction with false-negative mining and reinforced positive-class knowledge distillation
errINFORMATION PROCESSING & MANAGEMENT
IF6.9
err2024-01-01
err2
errOAAI
errZeng, Daojian; Zhu, Jianling; Chen, Hongting; Dai, Jianhua; Jiang, Lincheng
err分享
err收藏
学者 查看更多内容