Document Sub-structure in Neural Machine Translation
This addresses the issue of context modeling in machine translation for documents with predictable structures like biographies, though it is incremental as it adapts existing statistical MT ideas to neural MT.
The paper tackled the problem of neural machine translation ignoring document sub-structure by proposing methods to incorporate section topic information, resulting in improvements such as a 0.5 BLEU point gain for Chinese-English translation.
Current approaches to machine translation (MT) either translate sentences in isolation, disregarding the context they appear in, or model context at the level of the full document, without a notion of any internal structure the document may have. In this work we consider the fact that documents are rarely homogeneous blocks of text, but rather consist of parts covering different topics. Some documents, such as biographies and encyclopedia entries, have highly predictable, regular structures in which sections are characterised by different topics. We draw inspiration from Louis and Webber (2014) who use this information to improve statistical MT and transfer their proposal into the framework of neural MT. We compare two different methods of including information about the topic of the section within which each sentence is found: one using side constraints and the other using a cache-based model. We create and release the data on which we run our experiments - parallel corpora for three language pairs (Chinese-English, French-English, Bulgarian-English) from Wikipedia biographies, which we extract automatically, preserving the boundaries of sections within the articles.