Multilingual Multimodal Learning with Machine Translated Text

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Documents

  • Fulltext

    Final published version, 1.87 MB, PDF document

Most vision-and-language pretraining research focuses on English tasks. However, the creation of multilingual multimodal evaluation datasets (e.g. Multi30K, xGQA, XVNLI, and MaRVL) poses a new challenge in finding high-quality training data that is both multilingual and multimodal. In this paper, we investigate whether machine translating English multimodal data can be an effective proxy for the lack of readily available multilingual data. We call this framework TD-MML: Translated Data for Multilingual Multimodal Learning, and it can be applied to any multimodal dataset and model. We apply it to both pretraining and fine-tuning data with a state-of-the-art model. In order to prevent models from learning from low-quality translated text, we propose two metrics for automatically removing such translations from the resulting datasets. In experiments on five tasks across 20 languages in the IGLUE benchmark, we show that translated data can provide a useful signal for multilingual multimodal learning, both at pretraining and fine-tuning.
Original languageEnglish
Title of host publicationFindings of the Association for Computational Linguistics: EMNLP 2022
PublisherAssociation for Computational Linguistics (ACL)
Publication date2022
Pages4178–4193
Publication statusPublished - 2022
EventThe 2022 Conference on Empirical Methods in Natural Language Processing - Abu Dhabi, Abu Dhabi
Duration: 7 Dec 202211 Dec 2022
Conference number: 17
https://2022.emnlp.org/

Conference

ConferenceThe 2022 Conference on Empirical Methods in Natural Language Processing
Nummer17
LocationAbu Dhabi
ByAbu Dhabi
Periode07/12/202211/12/2022
Internetadresse

ID: 339327319