Sentence-Level Multilingual Multi-modal Embedding for Natural Language Processing

Iacer Calixto, Qun Liu


Abstract
We propose a novel discriminative ranking model that learns embeddings from multilingual and multi-modal data, meaning that our model can take advantage of images and descriptions in multiple languages to improve embedding quality. To that end, we introduce an objective function that uses pairwise ranking adapted to the case of three or more input sources. We compare our model against different baselines, and evaluate the robustness of our embeddings on image–sentence ranking (ISR), semantic textual similarity (STS), and neural machine translation (NMT). We find that the additional multilingual signals lead to improvements on all three tasks, and we highlight that our model can be used to consistently improve the adequacy of translations generated with NMT models when re-ranking n-best lists.
Anthology ID:
R17-1020
Volume:
Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017
Month:
September
Year:
2017
Address:
Varna, Bulgaria
Venue:
RANLP
SIG:
Publisher:
INCOMA Ltd.
Note:
Pages:
139–148
Language:
URL:
https://doi.org/10.26615/978-954-452-049-6_020
DOI:
10.26615/978-954-452-049-6_020
Bibkey:
Copy Citation:
PDF:
https://doi.org/10.26615/978-954-452-049-6_020