Machine Translation with Unsupervised Length-Constraints

Machine Translation with Unsupervised Length-Constraints Jan Niehues jan.niehues@maastrichtuniversity.nl Department of Data Science and Knowledge Engineering (DKE), Maastricht University, Maas- tricht, The Netherlands Abstract We have seen significant improvements in machine translation due to the usage of deep learn- ing. While the improvements in translation quality are impressive, the encoder-decoder archi- tecture enables many more possibilities. In this paper, we explore one of these, the generation of constrained translation. We focus on length constraints, which are essential if the translation should be displayed in a given format. In this work, we propose an end-to-end approach for this task. Compared to a traditional method that first translates and then performs sentence compression, the text compression is learned completely unsupervised. We address the challenge of data availability as well as inves- tigate several methods to integrate the constraints into the model. By combining the idea with zero-shot multilingual machine translation, we are also able to perform unsupervised monolin- gual sentence compression. Using the proposed approach, we are able to improve the translation quality for translation with length constraints as well as for monolingual length compression. In addition, the results are confirmed by a human evaluation. 1 Introduction Neural machine translation (NMT) (Sutskever et al., 2014; Bahdanau et al., 2014) exploits neu- ral networks to directly learn to transform sentences in a source language to sentences in a target language. This technique has significantly improved the quality of machine translation (Bojar et al., 2016; Cettolo et al., 2015). The advances in quality also allow for the application of this technology to new real-world applications. While research systems tend to purely focus on a high translation quality, real-world ap- plications often have additional requirements for the output of the system. One example is the mapping of markup information from the source text to the target text (Zenkel et al., 2019). In this work, we will focus on another use case, the generation of translations with given length constraints. Thereby, we focus on compression. That means the target length is shorter than the actual length of the translation. When translating from one language to another, the length of the source text is usually different from the length of the target text. While for most applications of machine translation this does not pose a problem, for some applications this significantly de- teriorates the user experience. For example, if the translation should be displayed in the same layout as the source text (e.g. in a website), it is advantageous if the length stays the same. An- other use case are captions for videos. A human is only capable of reading text up to a certain speed. For an optimal user experience, it is therefore not only important to present an accurate translation, but also to present the translation with a maximum number of words. Proceedings of the 14th Conference of the Association for Machine Translation in the Americas October 6 - 9, 2020, Volume 1: MT Research Track Page 21 A first approach to address this challenge would …

Content truncated to 500 words