计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (3): 97-106.doi: 10.11896/jsjkx.250500095

• 基于AGI技术的智能信息系统 • 上一篇    下一篇

多尺度聚合协同轴向语义引导的实体关系联合抽取方法

钱清1,3, 陈辉程1, 崔允贺2, 唐瑞雪1,3, 付金玫1   

  1. 1 贵州财经大学信息学院 贵阳 550025
    2 贵州大学计算机科学与技术学院文本计算与认知智能教育部工程研究中心 贵阳 550025
    3 贵州省算网融合全省重点实验室 贵阳 550025
  • 收稿日期:2025-05-22 修回日期:2025-09-12 发布日期:2026-03-12
  • 通讯作者: 崔允贺(yhcui@gzu.edu.cn)
  • 作者简介:(qqian2018_p@163.com)
  • 基金资助:
    国家自然科学基金(62462010,61902085);贵州省科技计划项目(黔科合基础-ZK[2022]一般018)

Joint Entity and Relation Extraction Method with Multi-scale Collaborative Aggregation and Axial-semantic Guidance

QIAN Qing1,3, CHEN Huicheng1, CUI Yunhe2, TANG Ruixue1,3, FU Jinmei1   

  1. 1 School of Information, Guizhou University of Finance and Economics, Guiyang 550025, China
    2 Engineering Research Center of Text Computing & Cognitive Intelligence, Ministry of Education, College of Computer Science & Technology, Guizhou University, Guiyang 550025, China
    3 Guizhou Provincial Key Laboratory of Computing and Network Convergence, Guiyang 550025, China
  • Received:2025-05-22 Revised:2025-09-12 Online:2026-03-12
  • About author:QIAN Qing,born in 1986,Ph.D,asso-ciate professor,is a member of CCF(No.K8203S).Her main research interests include natural language processing,digital media forensics and multimedia signal processing.
    CUl Yunhe,born in 1987,Ph.D,asso-ciate professor,is a member of CCF(No.F3600M).His main research interests include lightweight large mo-dels,network security,software-defined networks,data center networks and network telemetry.
  • Supported by:
    National Natural Science Foundation of China(62462010,61902085) and Guizhou Provincial Science and Technology Projects(QKH-Basic-ZK[2022]General 018).

PDF (PC)

137

摘要： 近年来,基于表填充的实体关系联合抽取方法取得了显著效果,但现有研究尚未考虑到词对间的边界关联性建模,以及构建词对语义相似性问题。为解决上述问题,提出了一种基于多尺度聚合协同轴向语义引导的实体关系联合抽取模型。首先,设计的多尺度语义聚合模块通过并行多个不同尺寸的深度卷积提取不同排列下词对间的边界关联信息,从而丰富词对语义,识别隐形实体。其次,轴向语义引导模块通过行列带状卷积从轴向上进行卷积注意力校准,强化词对关键语义表征,从而改善词对间语义相似问题。最后,在数据集NYT*,WebNLG*,NYT和WebNLG上进行实验,该方法分别取得了93.2%,94.5%,93.2%和91.4%的F1得分,相较于基线模型分别提高了0.1个百分点、0.6个百分点、0.4个百分点和1.0个百分点,表明其能够捕获词对边界关联以及精细化词对语义,提升了实体关系联合抽取的性能。

关键词: 自然语言处理, 实体关系联合抽取, 多尺度语义聚合, 轴向语义引导, 卷积注意力

Abstract: In recent years,table-filling approaches to joint entity-relation extraction have achieved impressive performance,yet they typically neglect two critical challenges:modelling boundary correlations among token pairs and distinguishing semantically similar token pairs.To address these gaps,this paper introduces a novel joint extraction model featuring multi-scale semantic aggregation and axial-semantic guidance.Firstly,multi-scale semantic aggregation module applies parallel depthwise convolutions of varying kernel sizes to capture boundary correlation information across multiple spatial arrangements,thereby enriching token-pair representations and facilitating the detection of implicit entities.Next,axial-semantic guidance module employsrow-and co-lumn-wise banded convolutions to perform axis-aligned attention calibration,strengthening key semantic features and effectively resolving high-similarity ambiguities.Comprehensive experiments on NYT*,WebNLG*,NYT,and WebNLG datasets yield F1 scores of 93.2%,94.5%,93.2%,and 91.4%－corresponding to absolute gains of 0.1 percentage points,0.6 percentage points,0.4 percentage points,and 1.0 percentage points over strong baselines.These results validate that explicitly capturing multi-scale boundary correlations and refining semantic alignment substantially enhances joint entity-relation extraction.

Key words: Natural language processing, Joint entity-relation extraction, Multi-scale semantic aggregation, Axial semantic gui-dance, Convolutional attention mechanism

中图分类号: 

  • TP391
[1]WANG Y H,MA F Y,WANG Y.Construction of a Fine-Grained Medical Knowledge Graph Based on Deep Learning[J].Computer Science,2024,51(S2):36-42.
[2]ZHANG Y,ZHANG J,CUI Z,et al.A Graph-based Relevance Matching Model for Ad-hoc Retrieval[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:4688-4696.
[3]WEI Z,SU J,WANG Y,et al.A Novel Cascade Binary Tagging Framework for Relational Triple Extraction[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.2020:1476-1488.
[4]SHANG Y M,HUANG H,SUN X,et al.Relational Triple Extraction:One Step is Enough[C]//Thirty-First International Joint Conference on Artificial Intelligence.2022:4360-4366.
[5]WADHWA S,AMIR S,WALLACE B.Revisiting Relation Extraction in the era of Large Language Models[C]//Proceedings of the 61st Annual Meeting of the Association for Computatio-nal Linguistics.2023:15566-15589.
[6]SHANG Y M,HUANG H,MAO X.OneRel:Joint Entity and Relation Extraction with One Module in One Step[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2022:11285-11293.
[7]DAI Q,YANG W,WANG L,et al.SOIRP:Subject-Object In-teraction and Reasoning Path based joint relational triple extraction by table filling[J].Neurocomputing,2024,580:127492.
[8]MIWA M,BANSAL M.End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics.2016:1105-1116.
[9]SUN K,ZHANG R,MENSAH S,et al.Recurrent Interaction Network for Jointly Extracting Entities and Classifying Relations[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing(EMNLP).2020:3722-3732.
[10]ZHAO X,DENG Y,YANG M,et al.A Comprehensive Survey on Relation Extraction:Recent Advances and New Frontiers[J].ACM Computing Surveys,2024,56(11):1-39.
[11]ZHAN Y,LI Z,ZHAO X,et al.A Simple Overlapping Relation Extraction Method Based on Dropout[C]//2022 International Joint Conference on Neural Networks(IJCNN).2022:1-8.
[12]ZHANG N,DENG S,YE H,et al.Robust triple extraction with cascade bidirectional capsule network[J].Expert Systems with Applications,2022,187:115806.
[13]WANG P,LIU X M,YANG G,et al.Joint extraction of entities and relationships based on subtext relationship enhancement[J].Computer Engineering and Design,2024,45(6):1780-1788.
[14]ZENG X,ZENG D,HE S,et al.Extracting Relational Facts by an End-to-End Neural Model with Copy Mechanism[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics.2018:506-514.
[15]SUI D,ZENG X,CHEN Y,et al.Joint Entity and Relation Extraction With Set Prediction Networks[J].IEEE Transactions on Neural Networks and Learning Systems,2020,35(9):12784-12795.
[16]ZHANG Z,YANG J,LIU H,et al.BTDM:A Bi-DirectionalTranslating Decoding Model-Based Relational Triple Extraction[J].Applied Sciences,2023,13(7):4447.
[17]GUO X R,WANG T J,YAN Y.Entity relationship extraction method based on local attention and local remote supervision[J].Journal of Jilin University(Engineering and Technology Edition),2025,55(1):307-315.
[18]WANG Y,YU B,ZHANG Y,et al.TPLinker:Single-stage Joint Extraction of Entities and Relations Through Token Pair Lin-king[C]//Proceedings of the 28th International Conference on Computational Linguistics.2020:1572-1582.
[19]NING J,YANG Z,SUN Y,et al.OD-RTE:A One-Stage Object Detection Framework for Relational Triple Extraction[C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics.2023:11120-11135.
[20]DEVLIN J,CHANG M W,LEE K,et al.BERT:Pre-training of Deep Bidirectional Transformers for Language Understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.2019:4171-4186.
[21]YAN H,SUN Y,LI X,et al.An Embarrassingly Easy butStrong Baseline for Nested Named Entity Recognition[C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics.2023:1442-1452.
[22]CAI X,LAI Q,WANG Y,et al.Poly Kernel Inception Network for Remote Sensing Detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and PatternRecognition(CVPR).2024:27706-27716.
[23]RIEDEL S,YAO L,MCCALLUM A.Modeling Relations and Their Mentions without Labeled Text[C]//Machine Learning and Knowledge Discovery in Databases.2010:148-163.
[24]GARDENT C,SHIMORINA A,NARAYAN S,et al.CreatingTraining Corpora for NLG Micro-Planners[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics.2017:179-188.
[25]KINGMA D P,BA J.Adam:A Method for Stochastic Optimization[J].arXiv:1412.6980,2014.
[26]ZHENG H,WEN R,CHEN X,et al.PRGC:Potential Relation and Global Correspondence Based Joint Relational Triple Extraction[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing.2021:6225-6235.
[27]WANG Z,WEI C,LI J,et al.CLFM:Contrastive Learning and Filter-attention Mechanism for Joint Relation Extraction[J].International Journal of Advanced Computer Science and Applications,2023,14(12):1004-1012.
[28]XIAO Y,ZHENG S,TIAN Z,et al.Bidirectional network-based relational triple extraction with prior relation mechanism[J].Knowledge and Information Systems,2024,67(1):427-453.
[29]ZHANG X,YU C,YAN R.ParTRE:A relational tripleextraction model of complicated entities and imbalanced relations in Parkinson’s disease[J].Journal of Biomedical Informatics,2024,152:104624.
[30]HU S,WANG H.DEL:a strategy for resolving redundancy inentity pairs within dual entity linker for relational triple extraction[J].Evolutionary Intelligence,2025,18(1):28.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发