@inproceedings{shokri-levitan-2023-gc,
title = "{GC}-Hunter at {I}mage{A}rg Shared Task: Multi-Modal Stance and Persuasiveness Learning",
author = "Shokri, Mohammad and
Levitan, Sarah Ita",
editor = "Alshomary, Milad and
Chen, Chung-Chi and
Muresan, Smaranda and
Park, Joonsuk and
Romberg, Julia",
booktitle = "Proceedings of the 10th Workshop on Argument Mining",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.argmining-1.17",
doi = "10.18653/v1/2023.argmining-1.17",
pages = "162--166",
abstract = "With the rising prominence of social media, users frequently supplement their written content with images. This trend has brought about new challenges in automatic processing of social media messages. In order to fully understand the meaning of a post, it is necessary to capture the relationship between the image and the text. In this work we address the two main objectives of the ImageArg shared task. Firstly, we aim to determine the stance of a multi-modal tweet toward a particular issue. We propose a strong baseline, fine-tuning transformer based models on concatenation of tweet text and image text. The second goal is to predict the impact of an image on the persuasiveness of the text in a multi-modal tweet. To capture the persuasiveness of an image, we train vision and language models on the data and explore other sets of features merged with the model, to enhance prediction power. Ultimately, both of these goals contribute toward the broader aim of understanding multi-modal messages on social media and how images and texts relate to each other.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shokri-levitan-2023-gc">
<titleInfo>
<title>GC-Hunter at ImageArg Shared Task: Multi-Modal Stance and Persuasiveness Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="family">Shokri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sarah</namePart>
<namePart type="given">Ita</namePart>
<namePart type="family">Levitan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 10th Workshop on Argument Mining</title>
</titleInfo>
<name type="personal">
<namePart type="given">Milad</namePart>
<namePart type="family">Alshomary</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chung-Chi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Smaranda</namePart>
<namePart type="family">Muresan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joonsuk</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">Romberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>With the rising prominence of social media, users frequently supplement their written content with images. This trend has brought about new challenges in automatic processing of social media messages. In order to fully understand the meaning of a post, it is necessary to capture the relationship between the image and the text. In this work we address the two main objectives of the ImageArg shared task. Firstly, we aim to determine the stance of a multi-modal tweet toward a particular issue. We propose a strong baseline, fine-tuning transformer based models on concatenation of tweet text and image text. The second goal is to predict the impact of an image on the persuasiveness of the text in a multi-modal tweet. To capture the persuasiveness of an image, we train vision and language models on the data and explore other sets of features merged with the model, to enhance prediction power. Ultimately, both of these goals contribute toward the broader aim of understanding multi-modal messages on social media and how images and texts relate to each other.</abstract>
<identifier type="citekey">shokri-levitan-2023-gc</identifier>
<identifier type="doi">10.18653/v1/2023.argmining-1.17</identifier>
<location>
<url>https://aclanthology.org/2023.argmining-1.17</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>162</start>
<end>166</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T GC-Hunter at ImageArg Shared Task: Multi-Modal Stance and Persuasiveness Learning
%A Shokri, Mohammad
%A Levitan, Sarah Ita
%Y Alshomary, Milad
%Y Chen, Chung-Chi
%Y Muresan, Smaranda
%Y Park, Joonsuk
%Y Romberg, Julia
%S Proceedings of the 10th Workshop on Argument Mining
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F shokri-levitan-2023-gc
%X With the rising prominence of social media, users frequently supplement their written content with images. This trend has brought about new challenges in automatic processing of social media messages. In order to fully understand the meaning of a post, it is necessary to capture the relationship between the image and the text. In this work we address the two main objectives of the ImageArg shared task. Firstly, we aim to determine the stance of a multi-modal tweet toward a particular issue. We propose a strong baseline, fine-tuning transformer based models on concatenation of tweet text and image text. The second goal is to predict the impact of an image on the persuasiveness of the text in a multi-modal tweet. To capture the persuasiveness of an image, we train vision and language models on the data and explore other sets of features merged with the model, to enhance prediction power. Ultimately, both of these goals contribute toward the broader aim of understanding multi-modal messages on social media and how images and texts relate to each other.
%R 10.18653/v1/2023.argmining-1.17
%U https://aclanthology.org/2023.argmining-1.17
%U https://doi.org/10.18653/v1/2023.argmining-1.17
%P 162-166
Markdown (Informal)
[GC-Hunter at ImageArg Shared Task: Multi-Modal Stance and Persuasiveness Learning](https://aclanthology.org/2023.argmining-1.17) (Shokri & Levitan, ArgMining-WS 2023)
ACL