We wrote a new paper to report the results of the WMT18 shared task on Quality Estimation, i.e. the task of predicting the quality of the output of machine translation systems, at various granularity levels: word, phrase, sentence and document. This year we include four language pairs, three text domains, and translations produced by both statistical and neural machine translation systems. Participating teams from ten institutions submitted a variety of systems to different task variants and language pairs.

Read the full paper

I will be presenting our findings in Brussels this Autumn, November 1st, during WMT'18.

UPDATE (13/12/2018) -- The slides of the presentation are now available online.