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

Jindřich Libovický; Zdeněk Kasner; Jindřich Helcl; Ondřej Dušek

doi:10.18653/v1/2020.ngt-1.18

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

Jindřich Libovický, Zdeněk Kasner, Jindřich Helcl, Ondřej Dušek

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract / Description of output

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.

Original language	English
Title of host publication	Proceedings of the Fourth Workshop on Neural Generation and Translation
Place of Publication	Online
Publisher	Association for Computational Linguistics
Pages	153-160
Number of pages	8
ISBN (Electronic)	978-1-952148-17-0
DOIs	https://doi.org/10.18653/v1/2020.ngt-1.18
Publication status	Published - 10 Jul 2020
Event	The 4th Workshop on Neural Generation and Translation - Online workshop, Seattle, United States Duration: 10 Jul 2020 → 10 Jul 2020 https://sites.google.com/view/wngt20

Workshop

Workshop	The 4th Workshop on Neural Generation and Translation
Abbreviated title	WNGT 2020
Country/Territory	United States
City	Seattle
Period	10/07/20 → 10/07/20
Internet address	https://sites.google.com/view/wngt20

Access to Document

10.18653/v1/2020.ngt-1.18Licence: Creative Commons: Attribution (CC-BY)

Expand and Filter_LIBOVICKY_DOA11052020_VOR_CC-BYFinal published version, 293 KBLicence: Creative Commons: Attribution (CC-BY)

https://www.aclweb.org/anthology/2020.ngt-1.18Licence: Creative Commons: Attribution (CC-BY)

Cite this

@inproceedings{742077190a754418a9653a6c34094ced,

title = "Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task",

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.",

author = "Jind{\v r}ich Libovick{\'y} and Zden{\v e}k Kasner and Jind{\v r}ich Helcl and Ond{\v r}ej Du{\v s}ek",

year = "2020",

month = jul,

day = "10",

doi = "10.18653/v1/2020.ngt-1.18",

language = "English",

pages = "153--160",

booktitle = "Proceedings of the Fourth Workshop on Neural Generation and Translation",

publisher = "Association for Computational Linguistics",

note = "The 4th Workshop on Neural Generation and Translation, WNGT 2020 ; Conference date: 10-07-2020 Through 10-07-2020",

url = "https://sites.google.com/view/wngt20",

}

Libovický, J, Kasner, Z, Helcl, J & Dušek, O 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. Association for Computational Linguistics, Online, pp. 153-160, The 4th Workshop on Neural Generation and Translation, Seattle, Washington, United States, 10/07/20. https://doi.org/10.18653/v1/2020.ngt-1.18

Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task. / Libovický, Jindřich; Kasner, Zdeněk; Helcl, Jindřich et al.
Proceedings of the Fourth Workshop on Neural Generation and Translation. Online: Association for Computational Linguistics, 2020. p. 153-160.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task

AU - Libovický, Jindřich

AU - Kasner, Zdeněk

AU - Helcl, Jindřich

AU - Dušek, Ondřej

PY - 2020/7/10

Y1 - 2020/7/10

N2 - 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.

AB - 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.

U2 - 10.18653/v1/2020.ngt-1.18

DO - 10.18653/v1/2020.ngt-1.18

M3 - Conference contribution

SP - 153

EP - 160

BT - Proceedings of the Fourth Workshop on Neural Generation and Translation

PB - Association for Computational Linguistics

CY - Online

T2 - The 4th Workshop on Neural Generation and Translation

Y2 - 10 July 2020 through 10 July 2020

ER -

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

Abstract / Description of output

Workshop

Access to Document

Fingerprint

Cite this