Reinforcement of BERT with Dependency-Parsing Based Attention Mask

被引:0
|
作者
Mechouma, Toufik [1 ]
Biskri, Ismail [2 ]
Meunier, Jean Guy [1 ]
机构
[1] Univ Quebec Montreal, Montreal, PQ, Canada
[2] Univ Quebec Trois Rivieres, Trois Rivieres, PQ, Canada
来源
ADVANCES IN COMPUTATIONAL COLLECTIVE INTELLIGENCE, ICCCI 2022 | 2022年 / 1653卷
关键词
Bert; Transformers; Attention mechanisms; Dependency parsing;
D O I
10.1007/978-3-031-16210-7_9
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Dot-Product based attention mechanism is among recent attention mechanisms. It showed an outstanding performance with BERT. In this paper, we propose a dependency-parsing mask to reinforce the padding mask, at the multi-head attention units. Padding mask, is already used to filter padding positions. The proposed mask, aims to improve BERT attention filter. The conducted experiments, show that BERT performs better with the proposed mask.
引用
收藏
页码:112 / 122
页数:11
相关论文
共 50 条
  • [31] Exploring Automatic Feature Selection for Transition-Based Dependency Parsing
    Ballesteros, Miguel
    PROCESAMIENTO DEL LENGUAJE NATURAL, 2013, (51): : 119 - 126
  • [32] A survey of syntactic-semantic parsing based on constituent and dependency structures
    MeiShan Zhang
    Science China Technological Sciences, 2020, 63 : 1898 - 1920
  • [33] Research on Methods of Microblogging Sentiment Feature Extraction Based on Dependency Parsing
    Li Yonggan
    Zhou Xueguang
    Guo Wei
    Zhang Huanguo
    PROCEEDINGS OF THE 2015 INTERNATIONAL CONFERENCE ON INTELLIGENT SYSTEMS RESEARCH AND MECHATRONICS ENGINEERING, 2015, 121 : 581 - 589
  • [34] Design and Implementation of Weibo Sentiment Analysis Based on LDA and Dependency Parsing
    Yonggan Li
    Xueguang Zhou
    Yan Sun
    Huanguo Zhang
    中国通信, 2016, 13 (11) : 91 - 105
  • [35] ADPG: Biomedical entity recognition based on Automatic Dependency Parsing Graph
    Yang, Yumeng
    Lin, Hongfei
    Yang, Zhihao
    Zhang, Yijia
    Zhao, Di
    Huai, Shuaiheng
    JOURNAL OF BIOMEDICAL INFORMATICS, 2023, 140
  • [36] Mining Feature-Opinion from Reviews Based on Dependency Parsing
    He, Tieke
    Hao, Rui
    Qi, Hang
    Liu, Jia
    Wu, Qing
    INTERNATIONAL JOURNAL OF SOFTWARE ENGINEERING AND KNOWLEDGE ENGINEERING, 2016, 26 (9-10) : 1581 - 1591
  • [37] Design and Implementation of Weibo Sentiment Analysis Based on LDA and Dependency Parsing
    Li, Yonggan
    Zhou, Xueguang
    Sun, Yan
    Zhang, Huanguo
    CHINA COMMUNICATIONS, 2016, 13 (11) : 91 - 105
  • [38] Combining Dependency Parsing and a Lexical Network Based on Lexical Functions for the Identification of Collocations
    Fonseca, Alexsandro
    Sadat, Fatiha
    Lareau, Francois
    COMPUTATIONAL AND CORPUS-BASED PHRASEOLOGY, EUROPHRAS 2017, 2017, 10596 : 447 - 461
  • [39] Deep Graph-Based Character-Level Chinese Dependency Parsing
    Wu, Linzhi
    Zhang, Meishan
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2021, 29 : 1329 - 1339
  • [40] A Multiscale Interactive Attention Short Text Classification Model Based on BERT
    Zhou, Lu
    Wang, Peng
    Zhang, Huijun
    Wu, Shengbo
    Zhang, Tao
    IEEE ACCESS, 2024, 12 : 160992 - 161001