The Memad Submission To The WMT18 Multimodal Translation Task

Grönroos Stig-arne, Huet Benoit, Kurimo Mikko, Laaksonen Jorma, Merialdo Bernard, Pham Phu, Sjöberg Mats, Sulubacak Umut, Tiedemann Jörg, Troncy Raphael, Vázquez Raúl. Arxiv 2018

[Paper]    
Applications Model Architecture Multimodal Models Pretraining Methods Reinforcement Learning Transformer

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.

Similar Work