In this paper, we propose an effective way for biasing the attention
mechanism of a sequence-to-sequence neural machine translation (NMT) model
towards the well-studied statistical word alignment models. We show that our
novel guided alignment training approach improves translation quality on
real-life e-commerce texts consisting of product titles and descriptions,
overcoming the problems posed by many unknown words and a large type/token
ratio. We also show that meta-data associated with input texts such as topic or
category information can significantly improve translation quality when used as
an additional signal to the decoder part of the network. With both novel
features, the BLEU score of the NMT system on a product title set improves from
18.6 to 21.3%. Even larger MT quality gains are obtained through domain
adaptation of a general domain NMT system to e-commerce data. The developed NMT
system also performs well on the IWSLT speech translation task, where an
ensemble of four variant systems outperforms the phrase-based baseline by 2.1%
BLEU absolute.

本文提出了一种有效的方法来偏置序列到序列神经机器翻译模型（NMT）的注意机制，使其趋向于研究过的词对齐模型。我们展示了我们的新型引导对齐训练方法如何在真实生活的电商文本翻译中提高了翻译质量，克服了许多未知单词和大型类型 / 令牌比。我们还表明，与输入文本相关的元数据（例如主题或类别信息）作为网络解码器部分的附加信号可显着提高翻译质量。有了这两个新功能，NMT 系统在一个产品标题集上的 BLEU 分数从 18.6 提高到 21.3％。通过将通用领域 NMT 系统进行领域适应也可获得更大的机器翻译质量提升。这个开发的 NMT 系统在 IWSLT 语音翻译任务中也表现良好，其中四个变体系统的集合比基线的基于短语的系统的 BLEU 分数提高了 2.1％。