The MeMAD submission to the WMT18 multimodal translation task

Stig-Arne Grönroos, Benoit Huet, Mikko Kurimo, Jorma Laaksonen, Bernard Merialdo, Phu Pham, Mats Sjöberg, Umut Sulubacak, Jörg Tiedemann, Raphael Troncy, Raúl Vázquez

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

47 Citations (Scopus)
65 Downloads (Pure)


This paper describes the MeMAD project entry to the WMT Multimodal Machine Translation Shared Task. We propose adapting the Transformer neural machine translation (NMT) architecture to a multi-modal setting. In this paper, we also describe the preliminary experiments with text-only translation systems leading us up to this choice. We have the top scoring system for both English-to-German and English-to-French, according to the automatic metrics for flickr18. Our experiments show that the effect of the visual features in our system is small. Our largest gains come from the quality of the underlying text-only NMT system. We find that appropriate use of additional data is effective.
Original languageEnglish
Title of host publicationProceedings of the Third Conference on Machine Translation (WMT), Volume 2: Shared Task Papers
PublisherAssociation for Computational Linguistics
Number of pages9
ISBN (Print)978-1-948087-81-0
Publication statusPublished - 2018
MoE publication typeA4 Conference publication
EventConference on Machine Translation - Brussels, Belgium
Duration: 31 Oct 20181 Dec 2018
Conference number: 3


ConferenceConference on Machine Translation
Abbreviated titleWMT


Dive into the research topics of 'The MeMAD submission to the WMT18 multimodal translation task'. Together they form a unique fingerprint.

Cite this