@inproceedings{rodriguez-bribiesca-etal-2021-multimodal,
title = "Multimodal Weighted Fusion of Transformers for Movie Genre Classification",
author = "Rodr{\'\i}guez Bribiesca, Isaac and
L{\'o}pez Monroy, Adri{\'a}n Pastor and
Montes-y-G{\'o}mez, Manuel",
editor = "Zadeh, Amir and
Morency, Louis-Philippe and
Liang, Paul Pu and
Ross, Candace and
Salakhutdinov, Ruslan and
Poria, Soujanya and
Cambria, Erik and
Shi, Kelly",
booktitle = "Proceedings of the Third Workshop on Multimodal Artificial Intelligence",
month = jun,
year = "2021",
address = "Mexico City, Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.maiworkshop-1.1",
doi = "10.18653/v1/2021.maiworkshop-1.1",
pages = "1--5",
abstract = "The Multimodal Transformer showed to be a competitive model for multimodal tasks involving textual, visual and audio signals. However, as more modalities are involved, its late fusion by concatenation starts to have a negative impact on the model{'}s performance. Besides, interpreting model{'}s predictions becomes difficult, as one would have to look at the different attention activation matrices. In order to overcome these shortcomings, we propose to perform late fusion by adding a GMU module, which effectively allows the model to weight modalities at instance level, improving its performance while providing a better interpretabilty mechanism. In the experiments, we compare our proposed model (MulT-GMU) against the original implementation (MulT-Concat) and a SOTA model tested in a movie genre classification dataset. Our approach, MulT-GMU, outperforms both, MulT-Concat and previous SOTA model.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rodriguez-bribiesca-etal-2021-multimodal">
<titleInfo>
<title>Multimodal Weighted Fusion of Transformers for Movie Genre Classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Isaac</namePart>
<namePart type="family">Rodríguez Bribiesca</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adrián</namePart>
<namePart type="given">Pastor</namePart>
<namePart type="family">López Monroy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manuel</namePart>
<namePart type="family">Montes-y-Gómez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Third Workshop on Multimodal Artificial Intelligence</title>
</titleInfo>
<name type="personal">
<namePart type="given">Amir</namePart>
<namePart type="family">Zadeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Louis-Philippe</namePart>
<namePart type="family">Morency</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="given">Pu</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Candace</namePart>
<namePart type="family">Ross</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruslan</namePart>
<namePart type="family">Salakhutdinov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Soujanya</namePart>
<namePart type="family">Poria</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Erik</namePart>
<namePart type="family">Cambria</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kelly</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mexico City, Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The Multimodal Transformer showed to be a competitive model for multimodal tasks involving textual, visual and audio signals. However, as more modalities are involved, its late fusion by concatenation starts to have a negative impact on the model’s performance. Besides, interpreting model’s predictions becomes difficult, as one would have to look at the different attention activation matrices. In order to overcome these shortcomings, we propose to perform late fusion by adding a GMU module, which effectively allows the model to weight modalities at instance level, improving its performance while providing a better interpretabilty mechanism. In the experiments, we compare our proposed model (MulT-GMU) against the original implementation (MulT-Concat) and a SOTA model tested in a movie genre classification dataset. Our approach, MulT-GMU, outperforms both, MulT-Concat and previous SOTA model.</abstract>
<identifier type="citekey">rodriguez-bribiesca-etal-2021-multimodal</identifier>
<identifier type="doi">10.18653/v1/2021.maiworkshop-1.1</identifier>
<location>
<url>https://aclanthology.org/2021.maiworkshop-1.1</url>
</location>
<part>
<date>2021-06</date>
<extent unit="page">
<start>1</start>
<end>5</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Multimodal Weighted Fusion of Transformers for Movie Genre Classification
%A Rodríguez Bribiesca, Isaac
%A López Monroy, Adrián Pastor
%A Montes-y-Gómez, Manuel
%Y Zadeh, Amir
%Y Morency, Louis-Philippe
%Y Liang, Paul Pu
%Y Ross, Candace
%Y Salakhutdinov, Ruslan
%Y Poria, Soujanya
%Y Cambria, Erik
%Y Shi, Kelly
%S Proceedings of the Third Workshop on Multimodal Artificial Intelligence
%D 2021
%8 June
%I Association for Computational Linguistics
%C Mexico City, Mexico
%F rodriguez-bribiesca-etal-2021-multimodal
%X The Multimodal Transformer showed to be a competitive model for multimodal tasks involving textual, visual and audio signals. However, as more modalities are involved, its late fusion by concatenation starts to have a negative impact on the model’s performance. Besides, interpreting model’s predictions becomes difficult, as one would have to look at the different attention activation matrices. In order to overcome these shortcomings, we propose to perform late fusion by adding a GMU module, which effectively allows the model to weight modalities at instance level, improving its performance while providing a better interpretabilty mechanism. In the experiments, we compare our proposed model (MulT-GMU) against the original implementation (MulT-Concat) and a SOTA model tested in a movie genre classification dataset. Our approach, MulT-GMU, outperforms both, MulT-Concat and previous SOTA model.
%R 10.18653/v1/2021.maiworkshop-1.1
%U https://aclanthology.org/2021.maiworkshop-1.1
%U https://doi.org/10.18653/v1/2021.maiworkshop-1.1
%P 1-5
Markdown (Informal)
[Multimodal Weighted Fusion of Transformers for Movie Genre Classification](https://aclanthology.org/2021.maiworkshop-1.1) (Rodríguez Bribiesca et al., maiworkshop 2021)
ACL