USAAR-DFKI - The Transference Architecture for English-German Automatic Post-Editing

Santanu Pal; Hongfei Xu; Nico Herbig; Antonio Krüger; Josef van Genabith

In: Proceedings of the Fourth Conference on Machine Translation. Conference on Machine Translation (WMT-2019), August 1-2, Florence, Italy, Pages 124-131, ACL, 8/2019.


In this paper we present an English--German Automatic Post-Editing (APE) system called transference, submitted to the APE Task organized at WMT 2019. Our transference model is based on a multi-encoder transformer architecture. Unlike previous approaches, it (i) uses a transformer encoder block for src, (ii) followed by a transformer decoder block, but without masking, for self-attention on mt, which effectively acts as second encoder combining src --> mt, and (iii) feeds this representation into a final decoder block generating pe. This model improves over the raw black-box neural machine translation system by 0.9 and 1.0 absolute BLEU points on the WMT 2019 APE development and test set. Our submission ranked 3rd, however compared to the two top systems, performance differences are not statistically significant.


Weitere Links

Transference_WMT19.pdf (pdf, 201 KB )

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz