Distilling Translations with Visual Awareness

Julia Ive, Pranava Madhyastha, Lucia Specia


Abstract
Previous work on multimodal machine translation has shown that visual information is only needed in very specific cases, for example in the presence of ambiguous words where the textual context is not sufficient. As a consequence, models tend to learn to ignore this information. We propose a translate-and-refine approach to this problem where images are only used by a second stage decoder. This approach is trained jointly to generate a good first draft translation and to improve over this draft by (i) making better use of the target language textual context (both left and right-side contexts) and (ii) making use of visual context. This approach leads to the state of the art results. Additionally, we show that it has the ability to recover from erroneous or missing words in the source language.
Anthology ID:
P19-1653
Volume:
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2019
Address:
Florence, Italy
Editors:
Anna Korhonen, David Traum, Lluís Màrquez
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6525–6538
Language:
URL:
https://aclanthology.org/P19-1653
DOI:
10.18653/v1/P19-1653
Bibkey:
Cite (ACL):
Julia Ive, Pranava Madhyastha, and Lucia Specia. 2019. Distilling Translations with Visual Awareness. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 6525–6538, Florence, Italy. Association for Computational Linguistics.
Cite (Informal):
Distilling Translations with Visual Awareness (Ive et al., ACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/P19-1653.pdf
Code
 ImperialNLP/MMT-Delib
Data
Flickr30kMulti30K