Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task

Jindřich Libovický, Zdeněk Kasner, Jindřich Helcl, Ondřej Dušek


Abstract
We present our submission to the Simultaneous Translation And Paraphrase for Language Education (STAPLE) challenge. We used a standard Transformer model for translation, with a crosslingual classifier predicting correct translations on the output n-best list. To increase the diversity of the outputs, we used additional data to train the translation model, and we trained a paraphrasing model based on the Levenshtein Transformer architecture to generate further synonymous translations. The paraphrasing results were again filtered using our classifier. While the use of additional data and our classifier filter were able to improve results, the paraphrasing model produced too many invalid outputs to further improve the output quality. Our model without the paraphrasing component finished in the middle of the field for the shared task, improving over the best baseline by a margin of 10-22 % weighted F1 absolute.
Anthology ID:
2020.ngt-1.18
Volume:
Proceedings of the Fourth Workshop on Neural Generation and Translation
Month:
July
Year:
2020
Address:
Online
Venues:
ACL | NGT | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
153–160
Language:
URL:
https://aclanthology.org/2020.ngt-1.18
DOI:
10.18653/v1/2020.ngt-1.18
Bibkey:
Cite (ACL):
Jindřich Libovický, Zdeněk Kasner, Jindřich Helcl, and Ondřej Dušek. 2020. Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task. In Proceedings of the Fourth Workshop on Neural Generation and Translation, pages 153–160, Online. Association for Computational Linguistics.
Cite (Informal):
Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task (Libovický et al., NGT 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.ngt-1.18.pdf
Video:
 http://slideslive.com/38929832
Data
Duolingo STAPLE Shared Task