What do character-level models learn about morphology? The case of dependency parsing

Clara Vania, Andreas Grivas, Adam Lopez


Abstract
When parsing morphologically-rich languages with neural models, it is beneficial to model input at the character level, and it has been claimed that this is because character-level models learn morphology. We test these claims by comparing character-level models to an oracle with access to explicit morphological analysis on twelve languages with varying morphological typologies. Our results highlight many strengths of character-level models, but also show that they are poor at disambiguating some words, particularly in the face of case syncretism. We then demonstrate that explicitly modeling morphological case improves our best model, showing that character-level models can benefit from targeted forms of explicit morphological modeling.
Anthology ID:
D18-1278
Volume:
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
Month:
October-November
Year:
2018
Address:
Brussels, Belgium
Editors:
Ellen Riloff, David Chiang, Julia Hockenmaier, Jun’ichi Tsujii
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2573–2583
Language:
URL:
https://aclanthology.org/D18-1278
DOI:
10.18653/v1/D18-1278
Bibkey:
Cite (ACL):
Clara Vania, Andreas Grivas, and Adam Lopez. 2018. What do character-level models learn about morphology? The case of dependency parsing. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2573–2583, Brussels, Belgium. Association for Computational Linguistics.
Cite (Informal):
What do character-level models learn about morphology? The case of dependency parsing (Vania et al., EMNLP 2018)
Copy Citation:
PDF:
https://aclanthology.org/D18-1278.pdf
Attachment:
 D18-1278.Attachment.zip