CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models

Yeshan Wang; Ilia Markov

doi:10.18653/v1/2024.arabicnlp-1.51

CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

Abstract

We present the CLTL system designed for the ArAIEval Shared Task 2024 on multimodal propagandistic memes classification in Arabic. The challenge was divided into three subtasks: identifying propagandistic content from textual modality of memes (subtask 2A), from visual modality of memes (subtask 2B), and in a multimodal scenario when both modalities are combined (subtask 2C). We explored various uni-modal transformer models for Arabic language processing (subtask 2A), visual models for image processing (subtask 2B), and concatenated text and image embeddings using the Multilayer Perceptron fusion module for multimodal propagandistic memes classification (subtask 2C). Our system achieved 77.96% for subtask 2A, 71.04% for subtask 2B, and 79.80% for subtask 2C, ranking 2nd, 1st, and 3rd on the leaderboard.

Original language	English
Title of host publication	Proceedings of The Second Arabic Natural Language Processing Conference
Editors	Nizar Habash, Houda Bouamor, Ramy Eskander, Nadi Tomeh, Ibrahim Abu Farha, Ahmed Abdelali, Samia Touileb, Injy Hamed, Yaser Onaizan, Bashar Alhafni, Wissam Antoun, Salam Khalifa, Hatem Haddad, Imed Zitouni, Badr AlKhamissi, Rawan Almatham, Khalil Mrini
Publisher	Association for Computational Linguistics (ACL)
Pages	501-506
Number of pages	6
ISBN (Electronic)	9798891761322
DOIs	https://doi.org/10.18653/v1/2024.arabicnlp-1.51
Publication status	Published - 2024
Event	2nd Arabic Natural Language Processing Conference, ArabicNLP 2024 - Bangkok, Thailand Duration: 16 Aug 2024 → …

Conference

Conference	2nd Arabic Natural Language Processing Conference, ArabicNLP 2024
Country/Territory	Thailand
City	Bangkok
Period	16/08/24 → …

Bibliographical note

Publisher Copyright:
©2024 Association for Computational Linguistics.

Access to Document

10.18653/v1/2024.arabicnlp-1.51Licence: CC BY

Persistent URL (handle)

Persistent link

Cite this

Wang, Y., & Markov, I. (2024). CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models. In N. Habash, H. Bouamor, R. Eskander, N. Tomeh, I. A. Farha, A. Abdelali, S. Touileb, I. Hamed, Y. Onaizan, B. Alhafni, W. Antoun, S. Khalifa, H. Haddad, I. Zitouni, B. AlKhamissi, R. Almatham, & K. Mrini (Eds.), Proceedings of The Second Arabic Natural Language Processing Conference (pp. 501-506). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2024.arabicnlp-1.51

Wang, Yeshan ; Markov, Ilia. / CLTL at ArAIEval Shared Task : Multimodal Propagandistic Memes Classification Using Transformer Models. Proceedings of The Second Arabic Natural Language Processing Conference. editor / Nizar Habash ; Houda Bouamor ; Ramy Eskander ; Nadi Tomeh ; Ibrahim Abu Farha ; Ahmed Abdelali ; Samia Touileb ; Injy Hamed ; Yaser Onaizan ; Bashar Alhafni ; Wissam Antoun ; Salam Khalifa ; Hatem Haddad ; Imed Zitouni ; Badr AlKhamissi ; Rawan Almatham ; Khalil Mrini. Association for Computational Linguistics (ACL), 2024. pp. 501-506

@inproceedings{0d4996cde5f4420a9e4d18b18f318ae2,

title = "CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models",

abstract = "We present the CLTL system designed for the ArAIEval Shared Task 2024 on multimodal propagandistic memes classification in Arabic. The challenge was divided into three subtasks: identifying propagandistic content from textual modality of memes (subtask 2A), from visual modality of memes (subtask 2B), and in a multimodal scenario when both modalities are combined (subtask 2C). We explored various uni-modal transformer models for Arabic language processing (subtask 2A), visual models for image processing (subtask 2B), and concatenated text and image embeddings using the Multilayer Perceptron fusion module for multimodal propagandistic memes classification (subtask 2C). Our system achieved 77.96% for subtask 2A, 71.04% for subtask 2B, and 79.80% for subtask 2C, ranking 2nd, 1st, and 3rd on the leaderboard.",

author = "Yeshan Wang and Ilia Markov",

note = "Publisher Copyright: {\textcopyright}2024 Association for Computational Linguistics.; 2nd Arabic Natural Language Processing Conference, ArabicNLP 2024 ; Conference date: 16-08-2024",

year = "2024",

doi = "10.18653/v1/2024.arabicnlp-1.51",

language = "English",

pages = "501--506",

editor = "Nizar Habash and Houda Bouamor and Ramy Eskander and Nadi Tomeh and Farha, {Ibrahim Abu} and Ahmed Abdelali and Samia Touileb and Injy Hamed and Yaser Onaizan and Bashar Alhafni and Wissam Antoun and Salam Khalifa and Hatem Haddad and Imed Zitouni and Badr AlKhamissi and Rawan Almatham and Khalil Mrini",

booktitle = "Proceedings of The Second Arabic Natural Language Processing Conference",

publisher = "Association for Computational Linguistics (ACL)",

address = "United States",

}

Wang, Y & Markov, I 2024, CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models. in N Habash, H Bouamor, R Eskander, N Tomeh, IA Farha, A Abdelali, S Touileb, I Hamed, Y Onaizan, B Alhafni, W Antoun, S Khalifa, H Haddad, I Zitouni, B AlKhamissi, R Almatham & K Mrini (eds), Proceedings of The Second Arabic Natural Language Processing Conference. Association for Computational Linguistics (ACL), pp. 501-506, 2nd Arabic Natural Language Processing Conference, ArabicNLP 2024, Bangkok, Thailand, 16/08/24. https://doi.org/10.18653/v1/2024.arabicnlp-1.51

CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models. / Wang, Yeshan; Markov, Ilia.
Proceedings of The Second Arabic Natural Language Processing Conference. ed. / Nizar Habash; Houda Bouamor; Ramy Eskander; Nadi Tomeh; Ibrahim Abu Farha; Ahmed Abdelali; Samia Touileb; Injy Hamed; Yaser Onaizan; Bashar Alhafni; Wissam Antoun; Salam Khalifa; Hatem Haddad; Imed Zitouni; Badr AlKhamissi; Rawan Almatham; Khalil Mrini. Association for Computational Linguistics (ACL), 2024. p. 501-506.

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - CLTL at ArAIEval Shared Task

T2 - 2nd Arabic Natural Language Processing Conference, ArabicNLP 2024

AU - Wang, Yeshan

AU - Markov, Ilia

PY - 2024

Y1 - 2024

N2 - We present the CLTL system designed for the ArAIEval Shared Task 2024 on multimodal propagandistic memes classification in Arabic. The challenge was divided into three subtasks: identifying propagandistic content from textual modality of memes (subtask 2A), from visual modality of memes (subtask 2B), and in a multimodal scenario when both modalities are combined (subtask 2C). We explored various uni-modal transformer models for Arabic language processing (subtask 2A), visual models for image processing (subtask 2B), and concatenated text and image embeddings using the Multilayer Perceptron fusion module for multimodal propagandistic memes classification (subtask 2C). Our system achieved 77.96% for subtask 2A, 71.04% for subtask 2B, and 79.80% for subtask 2C, ranking 2nd, 1st, and 3rd on the leaderboard.

AB - We present the CLTL system designed for the ArAIEval Shared Task 2024 on multimodal propagandistic memes classification in Arabic. The challenge was divided into three subtasks: identifying propagandistic content from textual modality of memes (subtask 2A), from visual modality of memes (subtask 2B), and in a multimodal scenario when both modalities are combined (subtask 2C). We explored various uni-modal transformer models for Arabic language processing (subtask 2A), visual models for image processing (subtask 2B), and concatenated text and image embeddings using the Multilayer Perceptron fusion module for multimodal propagandistic memes classification (subtask 2C). Our system achieved 77.96% for subtask 2A, 71.04% for subtask 2B, and 79.80% for subtask 2C, ranking 2nd, 1st, and 3rd on the leaderboard.

UR - http://www.scopus.com/inward/record.url?scp=85204285570&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85204285570&partnerID=8YFLogxK

UR - https://aclanthology.org/volumes/2024.arabicnlp-1/

U2 - 10.18653/v1/2024.arabicnlp-1.51

DO - 10.18653/v1/2024.arabicnlp-1.51

M3 - Conference contribution

AN - SCOPUS:85204285570

SP - 501

EP - 506

BT - Proceedings of The Second Arabic Natural Language Processing Conference

A2 - Habash, Nizar

A2 - Bouamor, Houda

A2 - Eskander, Ramy

A2 - Tomeh, Nadi

A2 - Farha, Ibrahim Abu

A2 - Abdelali, Ahmed

A2 - Touileb, Samia

A2 - Hamed, Injy

A2 - Onaizan, Yaser

A2 - Alhafni, Bashar

A2 - Antoun, Wissam

A2 - Khalifa, Salam

A2 - Haddad, Hatem

A2 - Zitouni, Imed

A2 - AlKhamissi, Badr

A2 - Almatham, Rawan

A2 - Mrini, Khalil

PB - Association for Computational Linguistics (ACL)

Y2 - 16 August 2024

ER -

Wang Y, Markov I. CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models. In Habash N, Bouamor H, Eskander R, Tomeh N, Farha IA, Abdelali A, Touileb S, Hamed I, Onaizan Y, Alhafni B, Antoun W, Khalifa S, Haddad H, Zitouni I, AlKhamissi B, Almatham R, Mrini K, editors, Proceedings of The Second Arabic Natural Language Processing Conference. Association for Computational Linguistics (ACL). 2024. p. 501-506 Epub 2024 Aug. doi: 10.18653/v1/2024.arabicnlp-1.51

CLTL at ArAIEval Shared Task: Multimodal Propagandistic Memes Classification Using Transformer Models

Abstract

Conference

Bibliographical note

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this