Efficient Low-Resource Neural Machine Translation with Reread and Feedback Mechanism

Author:

Yu Zhiqiang1,Yu Zhengtao1ORCID,Guo Junjun1,Huang Yuxin1,Wen Yonghua1

Affiliation:

1. Faculty of Information Engineering and Automation, Yunnan Key Laboratory of Artificial Intelligence, Kunming University of Science and Technology, Kunming, China

Abstract

How to utilize information sufficiently is a key problem in neural machine translation (NMT), which is effectively improved in rich-resource NMT by leveraging large-scale bilingual sentence pairs. However, for low-resource NMT, lack of bilingual sentence pairs results in poor translation performance; therefore, taking full advantage of global information in the encoding-decoding process is effective for low-resource NMT. In this article, we propose a novel reread-feedback NMT architecture (RFNMT) for using global information. Our architecture builds upon the improved sequence-to-sequence neural network and consists of a double-deck attention-based encoder-decoder framework. In our proposed architecture, the information generated by the first-pass encoding and decoding process flows to the second-pass encoding process for more sufficient parameters initialization and information use. Specifically, we first propose a “reread” mechanism to transfer the outputs of the first-pass encoder to the second-pass encoder, and then the output is used for the initialization of the second-pass encoder. Second, we propose a “feedback” mechanism that transfers the first-pass decoder’s outputs to a second-pass encoder via an important weight model and an improved gated recurrent unit (GRU). Experiments on multiple datasets show that our approach achieves significant improvements over state-of-the-art NMT systems, especially in low-resource settings.

Funder

National key Research and Development Plan Project

Yunnan High-Tech Industry Development Project

Natural Science Foundation of Yunnan Province

National Natural Science Foundation of China

Publisher

Association for Computing Machinery (ACM)

Subject

General Computer Science

Reference35 articles.

1. Mikel Artetxe Gorka Labaka Eneko Agirre and Kyunghyun Cho. 2017. Unsupervised Neural Machine Translation. arxiv:cs.CL/1710.11041 Mikel Artetxe Gorka Labaka Eneko Agirre and Kyunghyun Cho. 2017. Unsupervised Neural Machine Translation. arxiv:cs.CL/1710.11041

2. Dzmitry Bahdanau Kyunghyun Cho and Yoshua Bengio. 2014. Neural Machine Translation by Jointly Learning to Align and Translate. arxiv:cs.CL/1409.0473 Dzmitry Bahdanau Kyunghyun Cho and Yoshua Bengio. 2014. Neural Machine Translation by Jointly Learning to Align and Translate. arxiv:cs.CL/1409.0473

3. Franck Burlot and François Yvon. 2019. Using Monolingual Data in Neural Machine Translation: A Systematic Study. arxiv:cs.CL/1903.11437 Franck Burlot and François Yvon. 2019. Using Monolingual Data in Neural Machine Translation: A Systematic Study. arxiv:cs.CL/1903.11437

4. The FBK Participation in the WMT 2016 Automatic Post-editing Shared Task

Cited by 4 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. NLP-reliant Neural Machine Translation techniques used in smart city applications;Information System and Smart City;2023-10-02

2. Speech-to-speech Low-resource Translation;2023 IEEE 24th International Conference on Information Reuse and Integration for Data Science (IRI);2023-08

3. Improving Chinese-Vietnamese Neural Machine Translation with Linguistic Differences;ACM Transactions on Asian and Low-Resource Language Information Processing;2022-03-25

4. Improving thai-lao neural machine translation with similarity lexicon;Journal of Intelligent & Fuzzy Systems;2021-12-04

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3