Controlling the Output Length of Neural Machine Translation
This addresses a practical issue for users needing length-constrained translations, such as in media or document processing, but it is incremental as it builds on existing transformer architectures.
The paper tackles the problem of controlling output length in neural machine translation to match input length for applications like document translation and subtitles, showing that two methods—conditioning on length-ratio classes and enriching positional embeddings—can generate shorter translations and acquire interpretable linguistic skills.
The recent advances introduced by neural machine translation (NMT) are rapidly expanding the application fields of machine translation, as well as reshaping the quality level to be targeted. In particular, if translations have to fit some given layout, quality should not only be measured in terms of adequacy and fluency, but also length. Exemplary cases are the translation of document files, subtitles, and scripts for dubbing, where the output length should ideally be as close as possible to the length of the input text. This paper addresses for the first time, to the best of our knowledge, the problem of controlling the output length in NMT. We investigate two methods for biasing the output length with a transformer architecture: i) conditioning the output to a given target-source length-ratio class and ii) enriching the transformer positional embedding with length information. Our experiments show that both methods can induce the network to generate shorter translations, as well as acquiring interpretable linguistic skills.