Measuring and Increasing Context Usage in Context-Aware Machine Translation
This addresses the need for better evaluation and enhancement of context usage in document-level machine translation, offering incremental improvements for the field.
The paper tackled the problem of quantifying how much context-aware machine translation models actually use inter-sentential context, finding that target context is referenced more than source context and longer context has diminishing returns. They introduced a new training method, context-aware word dropout, which increased context usage and improved translation quality, with gains in BLEU, COMET, and performance on specific datasets.
Recent work in neural machine translation has demonstrated both the necessity and feasibility of using inter-sentential context -- context from sentences other than those currently being translated. However, while many current methods present model architectures that theoretically can use this extra context, it is often not clear how much they do actually utilize it at translation time. In this paper, we introduce a new metric, conditional cross-mutual information, to quantify the usage of context by these models. Using this metric, we measure how much document-level machine translation systems use particular varieties of context. We find that target context is referenced more than source context, and that conditioning on a longer context has a diminishing effect on results. We then introduce a new, simple training method, context-aware word dropout, to increase the usage of context by context-aware models. Experiments show that our method increases context usage and that this reflects on the translation quality according to metrics such as BLEU and COMET, as well as performance on anaphoric pronoun resolution and lexical cohesion contrastive datasets.