LIUM-CVC Submissions For WMT18 Multimodal Translation Task

Caglayan Ozan, Bardet Adrien, Bougares Fethi, Barrault Loïc, Wang Kai, Masana Marc, Herranz Luis, Van De Weijer Joost. Arxiv 2018

[Paper]    
Applications Attention Mechanism Model Architecture Multimodal Models

This paper describes the multimodal Neural Machine Translation systems developed by LIUM and CVC for WMT18 Shared Task on Multimodal Translation. This year we propose several modifications to our previous multimodal attention architecture in order to better integrate convolutional features and refine them using encoder-side information. Our final constrained submissions ranked first for English-French and second for English-German language pairs among the constrained submissions according to the automatic evaluation metric METEOR.

Similar Work