A BiLSTM-based System for Cross-lingual Pronoun Prediction

Sara Stymne, Sharid Loáiciga, Fabienne Cap


Abstract
We describe the Uppsala system for the 2017 DiscoMT shared task on cross-lingual pronoun prediction. The system is based on a lower layer of BiLSTMs reading the source and target sentences respectively. Classification is based on the BiLSTM representation of the source and target positions for the pronouns. In addition we enrich our system with dependency representations from an external parser and character representations of the source sentence. We show that these additions perform well for German and Spanish as source languages. Our system is competitive and is in first or second place for all language pairs.
Anthology ID:
W17-4805
Volume:
Proceedings of the Third Workshop on Discourse in Machine Translation
Month:
September
Year:
2017
Address:
Copenhagen, Denmark
Editors:
Bonnie Webber, Andrei Popescu-Belis, Jörg Tiedemann
Venue:
DiscoMT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
47–53
Language:
URL:
https://aclanthology.org/W17-4805
DOI:
10.18653/v1/W17-4805
Bibkey:
Cite (ACL):
Sara Stymne, Sharid Loáiciga, and Fabienne Cap. 2017. A BiLSTM-based System for Cross-lingual Pronoun Prediction. In Proceedings of the Third Workshop on Discourse in Machine Translation, pages 47–53, Copenhagen, Denmark. Association for Computational Linguistics.
Cite (Informal):
A BiLSTM-based System for Cross-lingual Pronoun Prediction (Stymne et al., DiscoMT 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-4805.pdf