Microsoft's Submission to the WMT2018 News Translation Task: How I Learned to Stop Worrying and Love the Data
This work addresses machine translation quality for English-German news text, though it is incremental as it builds on existing best practices.
The paper describes Microsoft's WMT2018 English-German news translation system, which achieved the highest BLEU score with a nearly 2-point margin over competitors and ranked first in human evaluation among constrained systems.
This paper describes the Microsoft submission to the WMT2018 news translation shared task. We participated in one language direction -- English-German. Our system follows current best-practice and combines state-of-the-art models with new data filtering (dual conditional cross-entropy filtering) and sentence weighting methods. We trained fairly standard Transformer-big models with an updated version of Edinburgh's training scheme for WMT2017 and experimented with different filtering schemes for Paracrawl. According to automatic metrics (BLEU) we reached the highest score for this subtask with a nearly 2 BLEU point margin over the next strongest system. Based on human evaluation we ranked first among constrained systems. We believe this is mostly caused by our data filtering/weighting regime.