Wait-info Policy: Balancing Source and Target at Information Level for Simultaneous Machine Translation
Shaolei Zhang 1,2, Shoutao Guo 1,2, Yang Feng 1,2 1Key Laboratory of Intelligent Information Processing Institute of Computing Technology, Chinese Academy of Sciences (ICT/CAS) 2 University of Chinese Academy of Sciences, Beijing, China {zhangshaolei20z, guoshoutao22z, Corresponding author: Yang Feng.
Abstract
Simultaneous machine translation (SiMT) outputs the translation while receiving the source inputs, and hence needs to balance the received source information and translated target information to make a reasonable decision between waiting for inputs or outputting translation. Previous methods always balance source and target information at the token level, either directly waiting for a fixed number of tokens or adjusting the waiting based on the current token. In this paper, we propose a Wait-info Policy to balance source and target at the information level. We first quantify the amount of information contained in each token, named info. Then during simultaneous translation, the decision of waiting or outputting is made based on the comparison results between the total info of previous target outputs and received source inputs. Experiments show that our method outperforms strong baselines under and achieves better balance via the proposed info111Code is available at https://github.com/ictnlp/Wait-info.
中文速览
同声传译（Simultaneous Machine Translation, SiMT）系统需要在还没读完源语言句子时就同步输出译文，关键难题在于如何合理权衡"已读源语信息"与"已译目标信息"来决定是继续等待还是立即输出。以往方法无论固定策略还是自适应策略，都把每个词等同对待、在词数层面做平衡，忽视了内容词比功能词信息量大得多这一事实。本文提出"Wait-info策略"，为每个词学习一个叫做"info"的标量来量化其信息量，决策时比较的是目标端已译词的信息总量与源端已读词的信息总量，当两者差距达到阈值K时才输出译文，从而在信息量层面而非词数层面做平衡；info值通过与注意力机制联合训练的无监督方式学得，无需额外标注数据。实验表明，该方法在多个语言对上均优于强基线，且能更有效地区分不同词的信息权重，在翻译质量与延迟的权衡上达到更好效果，对同声传译策略设计具有重要参考价值。
原文 arXiv:2210.11220;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2210.11220v1