@inproceedings{lee-etal-2022-efficient,
title = "Efficient Multilingual Multi-modal Pre-training through Triple Contrastive Loss",
author = "Lee, Youhan and
Lim, KyungTae and
Baek, Woonhyuk and
Roh, Byungseok and
Kim, Saehoon",
booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
month = oct,
year = "2022",
address = "Gyeongju, Republic of Korea",
publisher = "International Committee on Computational Linguistics",
url = "https://aclanthology.org/2022.coling-1.504",
pages = "5730--5744",
abstract = "Learning visual and textual representations in the shared space from web-scale image-text pairs improves the performance of diverse vision-and-language tasks, as well as modality-specific tasks. Many attempts in this framework have been made to connect English-only texts and images, and only a few works have been proposed to extend this framework in multilingual settings with the help of many translation pairs. In this multilingual approach, a typical setup is to use pairs of (image and English-text) and translation pairs. The major limitation of this approach is that the learning signal of aligning visual representation with under-resourced language representation is not strong, achieving a sub-optimal performance of vision-and-language tasks. In this work, we propose a simple yet effective enhancement scheme for previous multilingual multi-modal representation methods by using a limited number of pairs of images and non-English texts. In specific, our scheme fine-tunes a pre-trained multilingual model by minimizing a triplet contrastive loss on triplets of image and two different language texts with the same meaning, improving the connection between images and non-English texts. Experiments confirm that our enhancement strategy achieves performance gains in image-text retrieval, zero-shot image classification, and sentence embedding tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2022-efficient">
<titleInfo>
<title>Efficient Multilingual Multi-modal Pre-training through Triple Contrastive Loss</title>
</titleInfo>
<name type="personal">
<namePart type="given">Youhan</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">KyungTae</namePart>
<namePart type="family">Lim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Woonhyuk</namePart>
<namePart type="family">Baek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Byungseok</namePart>
<namePart type="family">Roh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Saehoon</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 29th International Conference on Computational Linguistics</title>
</titleInfo>
<originInfo>
<publisher>International Committee on Computational Linguistics</publisher>
<place>
<placeTerm type="text">Gyeongju, Republic of Korea</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Learning visual and textual representations in the shared space from web-scale image-text pairs improves the performance of diverse vision-and-language tasks, as well as modality-specific tasks. Many attempts in this framework have been made to connect English-only texts and images, and only a few works have been proposed to extend this framework in multilingual settings with the help of many translation pairs. In this multilingual approach, a typical setup is to use pairs of (image and English-text) and translation pairs. The major limitation of this approach is that the learning signal of aligning visual representation with under-resourced language representation is not strong, achieving a sub-optimal performance of vision-and-language tasks. In this work, we propose a simple yet effective enhancement scheme for previous multilingual multi-modal representation methods by using a limited number of pairs of images and non-English texts. In specific, our scheme fine-tunes a pre-trained multilingual model by minimizing a triplet contrastive loss on triplets of image and two different language texts with the same meaning, improving the connection between images and non-English texts. Experiments confirm that our enhancement strategy achieves performance gains in image-text retrieval, zero-shot image classification, and sentence embedding tasks.</abstract>
<identifier type="citekey">lee-etal-2022-efficient</identifier>
<location>
<url>https://aclanthology.org/2022.coling-1.504</url>
</location>
<part>
<date>2022-10</date>
<extent unit="page">
<start>5730</start>
<end>5744</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Efficient Multilingual Multi-modal Pre-training through Triple Contrastive Loss
%A Lee, Youhan
%A Lim, KyungTae
%A Baek, Woonhyuk
%A Roh, Byungseok
%A Kim, Saehoon
%S Proceedings of the 29th International Conference on Computational Linguistics
%D 2022
%8 October
%I International Committee on Computational Linguistics
%C Gyeongju, Republic of Korea
%F lee-etal-2022-efficient
%X Learning visual and textual representations in the shared space from web-scale image-text pairs improves the performance of diverse vision-and-language tasks, as well as modality-specific tasks. Many attempts in this framework have been made to connect English-only texts and images, and only a few works have been proposed to extend this framework in multilingual settings with the help of many translation pairs. In this multilingual approach, a typical setup is to use pairs of (image and English-text) and translation pairs. The major limitation of this approach is that the learning signal of aligning visual representation with under-resourced language representation is not strong, achieving a sub-optimal performance of vision-and-language tasks. In this work, we propose a simple yet effective enhancement scheme for previous multilingual multi-modal representation methods by using a limited number of pairs of images and non-English texts. In specific, our scheme fine-tunes a pre-trained multilingual model by minimizing a triplet contrastive loss on triplets of image and two different language texts with the same meaning, improving the connection between images and non-English texts. Experiments confirm that our enhancement strategy achieves performance gains in image-text retrieval, zero-shot image classification, and sentence embedding tasks.
%U https://aclanthology.org/2022.coling-1.504
%P 5730-5744
Markdown (Informal)
[Efficient Multilingual Multi-modal Pre-training through Triple Contrastive Loss](https://aclanthology.org/2022.coling-1.504) (Lee et al., COLING 2022)
ACL