《计算机应用》唯一官方网站

• •    下一篇

交互式机器翻译综述

廖兴滨1,2,秦小林1,2,张思齐1,2,钱杨舸1,2   

  1. 1. 中国科学院 成都计算机应用研究所
    2. 中国科学院大学 计算机科学与技术学院
  • 收稿日期:2021-12-06 修回日期:2022-05-03 发布日期:2022-06-30 出版日期:2022-06-30
  • 通讯作者: 秦小林
  • 作者简介:廖兴滨(1994—),男,湖北十堰人,硕士研究生, CCF会员(J9063G),主要研究方向:自然语言处理、数据泄露预防;秦小林(1980—),男,重庆合川人,研究员,博士生导师,博士,CCF会员,主要研究方向:人工智能、自动推理;张思齐(1995—),男,安徽淮南人,博士研究生,主要研究方向:自然语言处理、时序逻辑;钱杨舸(1997—),男,山东潍坊人,硕士研究生,主要研究方向:自然语言处理、人工智能。
  • 基金资助:
    四川省科技计划资助(2019ZDZX0006);中科院STS区域重点A类(KFJ-STS-QYZD-2021-21-001)

Review of interactive machine translation

LIAO Xingbin1,2, QIN Xiaolin1,2,ZHANG Siqi1,2,QIAN Yangge1,2   

  1. 1. Chengdu Institute of Computer Application, Chinese Academy of Sciences 2. School of Computer Science and Technology, University of Chinese Academy of Sciences
  • Received:2021-12-06 Revised:2022-05-03 Online:2022-06-30 Published:2022-06-30
  • About author:LIAO Xingbin, born in 1994, M. S. candidate. His research interests include natural language processing, data leakage prevention. QIN Xiaolin, born in 1980, Ph. D., research fellow. His research interests include artificial intelligence, automated reasoning. ZHANG Siqi, born in 1995, Ph. D. candidate. His research interests include natural language processing, temporal logic. QIAN Yangge, born in 1997,M. S. candidate. His research interests include natural language processing, artificial intelligence.
  • Supported by:
    Sichuan Science and Technology Program(2019ZDZX0006), Science and Technology Service Network Initiative (KFJ-STS-QYZD-2021-21-001)

摘要: 随着深度学习的发展和成熟,神经机器翻译的质量也越来越高,但仍不完美,为了达到可接受的翻译效果,需要人工进行后期编辑。交互式机器翻译是这种串行工作的一个替代,即在翻译过程中进行人工互动,由用户对翻译系统产生的候选翻译进行验证,或者由用户提供新的输入,系统根据用户当前的反馈生成新的候选译文,如此往复,直到产生一个使用户满意的输出。首先介绍交互式机器翻译的基本概念以及当前的研究进展;然后分类对一些常用方法和前沿工作加以介绍,简述每个工作的背景和创新之处;最后,探讨交互式机器翻译的发展趋势和研究难点。

关键词: 机器翻译;交互式机器翻译;交互式统计机器翻译;交互式神经机器翻译;强化学习方法, 自然语言处理

Abstract: Along with the development and maturation of deep learning, neural machine translations have obtained higher quality, but still imperfect, and Post-Editing by user is required to achieve acceptable translations. Interactive machine translation is an alternative to this serial work, i.e., human interaction is performed during the translation process, where the user verifies the candidate translation produced by the translation system and, if necessary, provides new input from the user until a satisfactory output is produced. Firstly, introduce the basic concept of interactive machine translation and the current research progress; then introduce some common methods and cutting-edge categorized works, while the background and innovation of each work are briefly described; finally, discuss some development trend and research difficulties of interactive machine translation.

Key words: machine translation, interactive machine translation, interactive statistical machine translation, interactive neural machine translation, reinforcement learning method, natural language processing

中图分类号: