1. Paper title
Simultaneous Translation Policies: From Fixed to Adaptive
2. link
https://www.aclweb.org/anthology/2020.acl-main.254.pdf
3. 摘要
Adaptive policies are better than fixed policies for simultaneous translation, since they can flexibly balance the tradeoff between translation quality and latency based on the current context information. But previous methods on obtaining adaptive policies either rely on complicated training process, or underperform simple fixed policies. We design an algorithm to achieve adaptive policies via a simple heuristic composition of a set of fixed policies. Experiments on Chinese→English and German→English show that our adaptive policies can outperform fixed ones by up to 4 BLEU points for the same latency, and more surprisingly, it even surpasses the BLEU score of full-sentence translation in the greedy mode (and very close to beam mode), but with much lower latency.
4. 要解决什么问题
Simultaneous translation(ST)的特点是：source sentence还没结束，翻译过程就已经开始。
ST相对于全句翻译的难点是：翻译质量与延时之间的权衡。
policy：翻译策略，决定每一步是READ还是WRITE。
已有的policy有：
- 把源分成几个trunk，用于SPEECH
- 一些简单的static rule，用于TEXT
- 根据已有信息动态计算，用于TEXT
已有的动态policy算法：
- 基于全句翻译模型，实时性不好
- 基于underlying翻译模型，翻译复杂
5. 作者的主要贡献
本文的动态policy算法： 将固定策略以一种简单启发式的方法组合，得到动态策略算法。
6. 得到了什么结果
- 实时翻译 Chinese→English and German→English 4 BLEU提升 在同样延时条件下
- 全句翻译 优于grid search，低于beam search