Z-Index at CheckThat! 2023: Unimodal and Multimodal Check-Worthiness Classification

Prerona Tarannum; Md Arid Hasan; Firoj Alam; Sheak Rashed Haider Noori

Z-Index at CheckThat! 2023: Unimodal and Multimodal Check-Worthiness Classification

Prerona Tarannum, Md Arid Hasan^*, Firoj Alam, Sheak Rashed Haider Noori

^*Corresponding author for this work

Qatar Computing Research Institute

Daffodil International University

Research output: Contribution to journal › Conference article › peer-review

1 Citation (Scopus)

Abstract

In this study, we report our participation in CheckThat! lab’s Task 1. The aim is to determine whether a claim made in either unimodal or multimodal content is worth fact-checking. We implemented standard preprocessing and fine-tuned the XLM-RoBERTa-large model. Additionally, we applied zero-shot learning and utilized a feed-forward network with embeddings for unimodal content. For subtask 1A submission, we used combined BERT-based models (BERT and BERT multilingual), ResNet50, and Feed Forward network and we ranked as 3^rd (Arabic) and 5^th (English). We used feed forward network with embeddings for subtask 1B submission and ranked as 3^rd in Arabic and 6^th in both English and Spanish. In further experiments, our evaluation shows that XLM-RoBERTa-large model outperforms the other models.

Original language	English
Pages (from-to)	482-493
Number of pages	12
Journal	CEUR Workshop Proceedings
Volume	3497
Publication status	Published - 2023
Event	24th Working Notes of the Conference and Labs of the Evaluation Forum, CLEF-WN 2023 - Thessaloniki, Greece Duration: 18 Sept 2023 → 21 Sept 2023

Keywords

Checkworthiness identification
Feed Forward Network
Misinformation
Multigenre Fact-checking
Multimodal Fact-checking
ResNet50
XLM-RoBERTa-large

Cite this

@article{e642795219664445b5b300c1f4e691fa,

title = "Z-Index at CheckThat! 2023: Unimodal and Multimodal Check-Worthiness Classification",

abstract = "In this study, we report our participation in CheckThat! lab{\textquoteright}s Task 1. The aim is to determine whether a claim made in either unimodal or multimodal content is worth fact-checking. We implemented standard preprocessing and fine-tuned the XLM-RoBERTa-large model. Additionally, we applied zero-shot learning and utilized a feed-forward network with embeddings for unimodal content. For subtask 1A submission, we used combined BERT-based models (BERT and BERT multilingual), ResNet50, and Feed Forward network and we ranked as 3rd (Arabic) and 5th (English). We used feed forward network with embeddings for subtask 1B submission and ranked as 3rd in Arabic and 6th in both English and Spanish. In further experiments, our evaluation shows that XLM-RoBERTa-large model outperforms the other models.",

keywords = "Checkworthiness identification, Feed Forward Network, Misinformation, Multigenre Fact-checking, Multimodal Fact-checking, ResNet50, XLM-RoBERTa-large",

author = "Prerona Tarannum and Hasan, \{Md Arid\} and Firoj Alam and Noori, \{Sheak Rashed Haider\}",

note = "Publisher Copyright: {\textcopyright} 2023 Copyright for this paper by its authors.; 24th Working Notes of the Conference and Labs of the Evaluation Forum, CLEF-WN 2023 ; Conference date: 18-09-2023 Through 21-09-2023",

year = "2023",

language = "English",

volume = "3497",

pages = "482--493",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

}

TY - JOUR

T1 - Z-Index at CheckThat! 2023

T2 - 24th Working Notes of the Conference and Labs of the Evaluation Forum, CLEF-WN 2023

AU - Tarannum, Prerona

AU - Hasan, Md Arid

AU - Alam, Firoj

AU - Noori, Sheak Rashed Haider

PY - 2023

Y1 - 2023

N2 - In this study, we report our participation in CheckThat! lab’s Task 1. The aim is to determine whether a claim made in either unimodal or multimodal content is worth fact-checking. We implemented standard preprocessing and fine-tuned the XLM-RoBERTa-large model. Additionally, we applied zero-shot learning and utilized a feed-forward network with embeddings for unimodal content. For subtask 1A submission, we used combined BERT-based models (BERT and BERT multilingual), ResNet50, and Feed Forward network and we ranked as 3rd (Arabic) and 5th (English). We used feed forward network with embeddings for subtask 1B submission and ranked as 3rd in Arabic and 6th in both English and Spanish. In further experiments, our evaluation shows that XLM-RoBERTa-large model outperforms the other models.

AB - In this study, we report our participation in CheckThat! lab’s Task 1. The aim is to determine whether a claim made in either unimodal or multimodal content is worth fact-checking. We implemented standard preprocessing and fine-tuned the XLM-RoBERTa-large model. Additionally, we applied zero-shot learning and utilized a feed-forward network with embeddings for unimodal content. For subtask 1A submission, we used combined BERT-based models (BERT and BERT multilingual), ResNet50, and Feed Forward network and we ranked as 3rd (Arabic) and 5th (English). We used feed forward network with embeddings for subtask 1B submission and ranked as 3rd in Arabic and 6th in both English and Spanish. In further experiments, our evaluation shows that XLM-RoBERTa-large model outperforms the other models.

KW - Checkworthiness identification

KW - Feed Forward Network

KW - Misinformation

KW - Multigenre Fact-checking

KW - Multimodal Fact-checking

KW - ResNet50

KW - XLM-RoBERTa-large

UR - https://www.scopus.com/pages/publications/85175642212

M3 - Conference article

AN - SCOPUS:85175642212

SN - 1613-0073

VL - 3497

SP - 482

EP - 493

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

Y2 - 18 September 2023 through 21 September 2023

ER -

Z-Index at CheckThat! 2023: Unimodal and Multimodal Check-Worthiness Classification

Abstract

Keywords

Other files and links

Fingerprint

Cite this