Mimicking Infants' Bilingual Language Acquisition for Domain Specialized Neural Machine Translation

被引:4
作者
Park, Chanjun [1 ,2 ]
Go, Woo-Young [3 ]
Eo, Sugyeong [1 ]
Moon, Hyeonseok [1 ]
Lee, Seolhwa [4 ]
Lim, Heuiseok [1 ]
机构
[1] Korea Univ, Dept Comp Sci & Engn, Seoul 02841, South Korea
[2] Upstage, Yongin 16942, South Korea
[3] Affiliated Inst ETRI, Future Strategy Ctr, Daejeon 34129, South Korea
[4] Univ Copenhagen, Dept Comp Sci, DK-1165 Copenhagen, Denmark
基金
新加坡国家研究基金会;
关键词
Training; Machine translation; Task analysis; Licenses; Computational modeling; Brain modeling; Terminology; Domain-specialized neural machine translation; cross communication method; deep learning; neural machine translation; SPLIT-ATTENTION; NETWORK MODEL; SPEECH; DISCRIMINATION; EXPERIENCE; PERCEPTION; JAPANESE; ADULTS;
D O I
10.1109/ACCESS.2022.3165572
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Existing methods of training domain-specialized neural machine translation (DS-NMT) models are based on the pretrain-finetuning approach (PFA). In this study, we reinterpret existing methods based on the perspective of cognitive science related to cross language speech perception. We propose the cross communication method (CCM), a new DS-NMT training approach. Inspired by the learning method of infants, we perform DS-NMT training by configuring and training DC and GC concurrently in batches. Quantitative and qualitative analysis of our experimental results show that CCM can achieve superior performance compared to the conventional methods. Additionally, we conducted an experiment considering the DS-NMT service to meet industrial demands.
引用
收藏
页码:38684 / 38693
页数:10
相关论文
共 40 条
  • [1] Online reading: a preliminary study of the impact of integrated and split-attention formats on L2 students' cognitive load
    Al-Shehri, Saleh
    Gitsaki, Christina
    [J]. RECALL, 2010, 22 : 356 - 375
  • [2] Bapna A, 2019, Arxiv, DOI [arXiv:1909.08478, 10.48550/arXiv.1909.08478]
  • [3] Brown C., 2000, 2 LANGUAGE ACQUISITI, P4
  • [4] Bilingual infants control their languages as they listen
    Byers-Heinlein, Krista
    Morin-Lessard, Elizabeth
    Lew-Williams, Casey
    [J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2017, 114 (34) : 9032 - 9037
  • [5] Cettolo Mauro, 2017, P 14 INT WORKSH SPOK, P2
  • [6] CHOMSKY Noam., 1965, Aspects of a theory of syntax, DOI DOI 10.21236/AD0616323
  • [7] Chu CH, 2018, Arxiv, DOI arXiv:1806.00258
  • [8] Clyne M., 2017, The handbook of sociolinguistics, P301, DOI [10.1002/9781405166256.ch18, DOI 10.1002/9781405166256.CH18]
  • [9] Dodge J, 2020, Arxiv, DOI [arXiv:2002.06305, 10.48550/arXiv.2002.06305, DOI 10.48550/ARXIV.2002.06305]
  • [10] Dissociable processes for learning the surface structure and abstract structure of sensorimotor sequences
    Dominey, PF
    Lelekov, TS
    [J]. JOURNAL OF COGNITIVE NEUROSCIENCE, 1998, 10 (06) : 734 - 751