The MeMAD Submission to the WMT18 Multimodal Translation Task

Stig-Arne Grönroos; Benoit Huet; Mikko Kurimo; Jorma Laaksonen; Bernard Merialdo; Phu Pham; Mats Sjöberg; Umut Sulubacak; Jörg Tiedemann; Raphaël Troncy; Juan Raúl Vázquez Carrillo

doi:10.18653/v1/W18-6439

The MeMAD Submission to the WMT18 Multimodal Translation Task

Stig-Arne Grönroos, Benoit Huet, Mikko Kurimo, Jorma Laaksonen, Bernard Merialdo, Phu Pham, Mats Sjöberg, Umut Sulubacak, Jörg Tiedemann, Raphaël Troncy, Juan Raúl Vázquez Carrillo

Tutkimustuotos: Artikkeli kirjassa/raportissa/konferenssijulkaisussa › Konferenssiartikkeli › Tieteellinen › vertaisarvioitu

Abstrakti

This paper describes the MeMAD project entry to the WMT Multimodal Machine
Translation Shared Task.

We propose adapting the Transformer neural machine translation (NMT) architecture to a multi-modal setting. In this paper, we also describe the preliminary experiments with text-only translation systems leading us up to this choice.

We have the top scoring system for both English-to-German and English-to-French, according to the automatic metrics for flickr18.

Our experiments show that the effect of the visual features in our system is small. Our largest gains come from the quality of the underlying text-only NMT system. We find that appropriate use of additional data is effective.

Alkuperäiskieli	englanti
Otsikko	Proceedings of the Third Conference on Machine Translation (WMT) : Shared Task Papers
Toimittajat	Ondřej Bojar, Rajen Chatterjee, Christian Federmann, Mark Fishel, Yvette Graham, Barry Haddow, Matthias Huck, Antonio Jimeno Yepes, Philipp Koehn, Christof Monz, Matteo Negri, Aurélie Névéol, Mariana Neves, Matt Post, Lucia Specia, Marco Turchi, Karin Verspoor
Sivumäärä	9
Julkaisupaikka	Stroudsburg
Kustantaja	The Association for Computational Linguistics
Julkaisupäivä	1 marrask. 2018
Sivut	603-611
ISBN (elektroninen)	978-1-948087-81-0
DOI - pysyväislinkit	https://doi.org/10.18653/v1/W18-6439
Tila	Julkaistu - 1 marrask. 2018
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisuussa
Tapahtuma	Conference on Machine Translation - 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP), Brussels, Belgia Kesto: 31 lokak. 2018 → 1 marrask. 2018 Konferenssinumero: 3 http://www.statmt.org/wmt18/

Tieteenalat

113 Tietojenkäsittely- ja informaatiotieteet
6121 Kielitieteet

Pääsy asiakirjaan

10.18653/v1/W18-6439Lisenssi: CC BY

Full textLopullinen julkaistu versio, 1,91 MBLisenssi: Määrittelemätön

MeMAD: Methods for Managing Audiovisual Data: Combining Automatic Efficiency with Human Accuracy
Hirvonen, M. (Principal Investigator), Tiedemann, J. (Osallistuja), Tiittula, L. (Osallistuja), Sulubacak, U. (Osallistuja), Vazquez , R. (Osallistuja) & Koponen, M. (Osallistuja)
European Commission / Horizon 2020
01/01/2018 → 31/03/2021
Projekti: Tutkimusprojekti

Siteeraa tätä

Grönroos, S.-A., Huet, B., Kurimo, M., Laaksonen, J., Merialdo, B., Pham, P., Sjöberg, M., Sulubacak, U., Tiedemann, J., Troncy, R., & Vázquez Carrillo, J. R. (2018). The MeMAD Submission to the WMT18 Multimodal Translation Task. teoksessa O. Bojar, R. Chatterjee, C. Federmann, M. Fishel, Y. Graham, B. Haddow, M. Huck, A. J. Yepes, P. Koehn, C. Monz, M. Negri, A. Névéol, M. Neves, M. Post, L. Specia, M. Turchi, & K. Verspoor (Toimittajat), Proceedings of the Third Conference on Machine Translation (WMT): Shared Task Papers (Sivut 603-611). The Association for Computational Linguistics. https://doi.org/10.18653/v1/W18-6439

Grönroos, Stig-Arne ; Huet, Benoit ; Kurimo, Mikko et al. / The MeMAD Submission to the WMT18 Multimodal Translation Task. Proceedings of the Third Conference on Machine Translation (WMT): Shared Task Papers. Toimittaja / Ondřej Bojar ; Rajen Chatterjee ; Christian Federmann ; Mark Fishel ; Yvette Graham ; Barry Haddow ; Matthias Huck ; Antonio Jimeno Yepes ; Philipp Koehn ; Christof Monz ; Matteo Negri ; Aurélie Névéol ; Mariana Neves ; Matt Post ; Lucia Specia ; Marco Turchi ; Karin Verspoor. Stroudsburg : The Association for Computational Linguistics, 2018. Sivut 603-611

@inproceedings{d755e4010d364e62aedbf0eab2f2ee8a,

title = "The MeMAD Submission to the WMT18 Multimodal Translation Task",

abstract = "This paper describes the MeMAD project entry to the WMT Multimodal MachineTranslation Shared Task.We propose adapting the Transformer neural machine translation (NMT) architecture to a multi-modal setting. In this paper, we also describe the preliminary experiments with text-only translation systems leading us up to this choice.We have the top scoring system for both English-to-German and English-to-French, according to the automatic metrics for flickr18.Our experiments show that the effect of the visual features in our system is small. Our largest gains come from the quality of the underlying text-only NMT system. We find that appropriate use of additional data is effective.",

keywords = "113 Computer and information sciences, 6121 Languages",

author = "Stig-Arne Gr{\"o}nroos and Benoit Huet and Mikko Kurimo and Jorma Laaksonen and Bernard Merialdo and Phu Pham and Mats Sj{\"o}berg and Umut Sulubacak and J{\"o}rg Tiedemann and Rapha{\"e}l Troncy and {V{\'a}zquez Carrillo}, {Juan Ra{\'u}l}",

year = "2018",

month = nov,

day = "1",

doi = "10.18653/v1/W18-6439",

language = "English",

pages = "603--611",

editor = "Ond{\v r}ej Bojar and Rajen Chatterjee and Christian Federmann and Mark Fishel and Yvette Graham and Barry Haddow and Matthias Huck and Yepes, {Antonio Jimeno} and Philipp Koehn and Christof Monz and Matteo Negri and Aur{\'e}lie N{\'e}v{\'e}ol and Mariana Neves and Matt Post and Lucia Specia and Marco Turchi and Karin Verspoor",

booktitle = "Proceedings of the Third Conference on Machine Translation (WMT)",

publisher = "The Association for Computational Linguistics",

address = "United States",

note = "Conference on Machine Translation, WMT18 ; Conference date: 31-10-2018 Through 01-11-2018",

url = "http://www.statmt.org/wmt18/",

}

Grönroos, S-A, Huet, B, Kurimo, M, Laaksonen, J, Merialdo, B, Pham, P, Sjöberg, M, Sulubacak, U, Tiedemann, J, Troncy, R & Vázquez Carrillo, JR 2018, The MeMAD Submission to the WMT18 Multimodal Translation Task. julkaisussa O Bojar, R Chatterjee, C Federmann, M Fishel, Y Graham, B Haddow, M Huck, AJ Yepes, P Koehn, C Monz, M Negri, A Névéol, M Neves, M Post, L Specia, M Turchi & K Verspoor (toim), Proceedings of the Third Conference on Machine Translation (WMT): Shared Task Papers. The Association for Computational Linguistics, Stroudsburg, Sivut 603-611, Conference on Machine Translation, Brussels, Belgia, 31/10/2018. https://doi.org/10.18653/v1/W18-6439

The MeMAD Submission to the WMT18 Multimodal Translation Task. / Grönroos, Stig-Arne; Huet, Benoit; Kurimo, Mikko et al.
Proceedings of the Third Conference on Machine Translation (WMT): Shared Task Papers. toim. / Ondřej Bojar; Rajen Chatterjee; Christian Federmann; Mark Fishel; Yvette Graham; Barry Haddow; Matthias Huck; Antonio Jimeno Yepes; Philipp Koehn; Christof Monz; Matteo Negri; Aurélie Névéol; Mariana Neves; Matt Post; Lucia Specia; Marco Turchi; Karin Verspoor. Stroudsburg: The Association for Computational Linguistics, 2018. s. 603-611.

Tutkimustuotos: Artikkeli kirjassa/raportissa/konferenssijulkaisussa › Konferenssiartikkeli › Tieteellinen › vertaisarvioitu

TY - GEN

T1 - The MeMAD Submission to the WMT18 Multimodal Translation Task

AU - Grönroos, Stig-Arne

AU - Huet, Benoit

AU - Kurimo, Mikko

AU - Laaksonen, Jorma

AU - Merialdo, Bernard

AU - Pham, Phu

AU - Sjöberg, Mats

AU - Sulubacak, Umut

AU - Tiedemann, Jörg

AU - Troncy, Raphaël

AU - Vázquez Carrillo, Juan Raúl

N1 - Conference code: 3

PY - 2018/11/1

Y1 - 2018/11/1

N2 - This paper describes the MeMAD project entry to the WMT Multimodal MachineTranslation Shared Task.We propose adapting the Transformer neural machine translation (NMT) architecture to a multi-modal setting. In this paper, we also describe the preliminary experiments with text-only translation systems leading us up to this choice.We have the top scoring system for both English-to-German and English-to-French, according to the automatic metrics for flickr18.Our experiments show that the effect of the visual features in our system is small. Our largest gains come from the quality of the underlying text-only NMT system. We find that appropriate use of additional data is effective.

AB - This paper describes the MeMAD project entry to the WMT Multimodal MachineTranslation Shared Task.We propose adapting the Transformer neural machine translation (NMT) architecture to a multi-modal setting. In this paper, we also describe the preliminary experiments with text-only translation systems leading us up to this choice.We have the top scoring system for both English-to-German and English-to-French, according to the automatic metrics for flickr18.Our experiments show that the effect of the visual features in our system is small. Our largest gains come from the quality of the underlying text-only NMT system. We find that appropriate use of additional data is effective.

KW - 113 Computer and information sciences

KW - 6121 Languages

U2 - 10.18653/v1/W18-6439

DO - 10.18653/v1/W18-6439

M3 - Conference contribution

SP - 603

EP - 611

BT - Proceedings of the Third Conference on Machine Translation (WMT)

A2 - Bojar, Ondřej

A2 - Chatterjee, Rajen

A2 - Federmann, Christian

A2 - Fishel, Mark

A2 - Graham, Yvette

A2 - Haddow, Barry

A2 - Huck, Matthias

A2 - Yepes, Antonio Jimeno

A2 - Koehn, Philipp

A2 - Monz, Christof

A2 - Negri, Matteo

A2 - Névéol, Aurélie

A2 - Neves, Mariana

A2 - Post, Matt

A2 - Specia, Lucia

A2 - Turchi, Marco

A2 - Verspoor, Karin

PB - The Association for Computational Linguistics

CY - Stroudsburg

T2 - Conference on Machine Translation

Y2 - 31 October 2018 through 1 November 2018

ER -

Grönroos SA, Huet B, Kurimo M, Laaksonen J, Merialdo B, Pham P et al. The MeMAD Submission to the WMT18 Multimodal Translation Task. julkaisussa Bojar O, Chatterjee R, Federmann C, Fishel M, Graham Y, Haddow B, Huck M, Yepes AJ, Koehn P, Monz C, Negri M, Névéol A, Neves M, Post M, Specia L, Turchi M, Verspoor K, toimittajat, Proceedings of the Third Conference on Machine Translation (WMT): Shared Task Papers. Stroudsburg: The Association for Computational Linguistics. 2018. s. 603-611 doi: 10.18653/v1/W18-6439

The MeMAD Submission to the WMT18 Multimodal Translation Task

Abstrakti

Tieteenalat

Pääsy asiakirjaan

Projektit

MeMAD: Methods for Managing Audiovisual Data: Combining Automatic Efficiency with Human Accuracy

Siteeraa tätä