Machine Translation Quality Estimation Across Domains

Machine Translation (MT) Quality Estimation (QE) aims to automatically measure the quality of MT system output without reference translations. In spite of the progress achieved in recent years, current MT QE systems are not capable of dealing with data coming from different train/test distributions or domains, and scenarios in which training data is scarce. We investigate different multitask learning methods that can cope with such limitations and show that they overcome current state-of-the-art methods in real-world conditions where training and test data come from different domains.