Pre-training neural machine translation with alignment information via optimal transport

被引:0
|
作者
Su, Xueping [1 ]
Zhao, Xingkai [1 ]
Ren, Jie [1 ]
Li, Yunhong [1 ]
Raetsch, Matthias [2 ]
机构
[1] Xian Polytech Univ, Sch Elect & Informat, Xian, Peoples R China
[2] Reutlingen Univ, Dept Engn, Interact & Mobile Robot & Artificial Intelligence, Reutlingen, Germany
基金
中国国家自然科学基金;
关键词
Optimal Transport; Alignment Information; Pre-training; Neural Machine Translation;
D O I
10.1007/s11042-023-17479-z
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
With the rapid development of globalization, the demand for translation between different languages is also increasing. Although pre-training has achieved excellent results in neural machine translation, the existing neural machine translation has almost no high-quality suitable for specific fields. Alignment information, so this paper proposes a pre-training neural machine translation with alignment information via optimal transport. First, this paper narrows the representation gap between different languages by using OTAP to generate domain-specific data for information alignment, and learns richer semantic information. Secondly, this paper proposes a lightweight model DR-Reformer, which uses Reformer as the backbone network, adds Dropout layers and Reduction layers, reduces model parameters without losing accuracy, and improves computational efficiency. Experiments on the Chinese and English datasets of AI Challenger 2018 and WMT-17 show that the proposed algorithm has better performance than existing algorithms.
引用
收藏
页码:48377 / 48397
页数:21
相关论文
共 50 条
  • [41] Roles of pre-training in deep neural networks from information theoretical perspective
    Furusho, Yasutaka
    Kubo, Takatomi
    Ikeda, Kazushi
    NEUROCOMPUTING, 2017, 248 : 76 - 79
  • [42] GENET: Unleashing the Power of Side Information for Recommendation via Hypergraph Pre-training
    Li, Yang
    Zhao, Qi'ao
    Lin, Chen
    Zhang, Zhenjie
    Zhu, Xiaomin
    Su, Jinsong
    DATABASE SYSTEMS FOR ADVANCED APPLICATIONS, DASFAA 2024, PT 3, 2025, 14852 : 343 - 352
  • [43] Pre-training on dynamic graph neural networks
    Chen, Ke-Jia
    Zhang, Jiajun
    Jiang, Linpu
    Wang, Yunyun
    Dai, Yuxuan
    NEUROCOMPUTING, 2022, 500 : 679 - 687
  • [44] Neural speech enhancement with unsupervised pre-training and mixture training
    Hao, Xiang
    Xu, Chenglin
    Xie, Lei
    NEURAL NETWORKS, 2023, 158 : 216 - 227
  • [45] SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint
    Sheng, Zhonghao
    Song, Kaitao
    Tan, Xu
    Ren, Yi
    Ye, Wei
    Zhang, Shikun
    Qin, Tao
    THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 13798 - 13805
  • [46] Neural Graph Matching for Pre-training Graph Neural Networks
    Hou, Yupeng
    Hu, Binbin
    Zhao, Wayne Xin
    Zhang, Zhiqiang
    Zhou, Jun
    Wen, Ji-Rong
    PROCEEDINGS OF THE 2022 SIAM INTERNATIONAL CONFERENCE ON DATA MINING, SDM, 2022, : 172 - 180
  • [47] Bilingual Mutual Information Based Adaptive Training for Neural Machine Translation
    Xu, Yangyifan
    Liu, Yijin
    Meng, Fandong
    Zhang, Jiajun
    Xu, Jinan
    Zhou, Jie
    ACL-IJCNLP 2021: THE 59TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS AND THE 11TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING, VOL 2, 2021, : 511 - 516
  • [48] Improved OOD Generalization via Adversarial Training and Pre-training
    Yi, Mingyangi
    Hou, Lu
    Sun, Jiacheng
    Shang, Lifeng
    Jiang, Xin
    Liu, Qun
    Ma, Zhi-Ming
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 139, 2021, 139
  • [49] On the Word Alignment from Neural Machine Translation
    Li, Xintong
    Li, Guanlin
    Liu, Lemao
    Meng, Max
    Shi, Shuming
    57TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2019), 2019, : 1293 - 1303
  • [50] Neural Machine Translation With Explicit Phrase Alignment
    Zhang, Jiacheng
    Luan, Huanbo
    Sun, Maosong
    Zhai, Feifei
    Xu, Jingfang
    Liu, Yang
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2021, 29 : 1001 - 1010