Attaining the Unattainable? Reassessing Claims of Human Parity in Neural Machine Translation.

We reassess a recent study (Hassan et al.,2018) that claimed that machine translation(MT) has reached human parity for the translation of news from Chinese into English, usingpairwise ranking and considering three variables that were not taken into account in thatprevious study: the language in which thesource side of the test set was originally written, the translation proficiency of the evaluators, and the provision of inter-sentential context. If we consider only original source text(i.e. not translated from another language, ortranslationese), then we find evidence showingthat human parity has not been achieved. Wecompare the judgments of professional translators against those of non-experts and discover that those of the experts result in higherinter-annotator agreement and better discrimination between human and machine translations. In addition, we analyse the human translations of the test set and identify importanttranslation issues. Finally, based on these findings, we provide a set of recommendations forfuture human evaluations of MT.