@inproceedings{alwajih-etal-2025-pearl,
title = "Pearl: A Multimodal Culturally-Aware {A}rabic Instruction Dataset",
author = "Alwajih, Fakhraddin and
Magdy, Samar M. and
El Mekki, Abdellah and
Nacar, Omer and
Nafea, Youssef and
Abdelfadil, Safaa Taher and
Yahya, Abdulfattah Mohammed and
Luqman, Hamzah and
Almarwani, Nada and
Aloufi, Samah and
Qawasmeh, Baraah and
Atou, Houdaifa and
Sibaee, Serry and
Alsayadi, Hamzah A. and
Al-Dhabyani, Walid and
Al-shaibani, Maged S. and
El aatar, Aya and
Qandos, Nour and
Alhamouri, Rahaf and
Ahmad, Samar and
AL-Ghrawi, Mohammed Anwar and
Yacoub, Aminetou and
AbuHweidi, Ruwa and
Lemin, Vatimetou Mohamed and
Abdel-Salam, Reem and
Bashiti, Ahlam and
Ammar, Adel and
Alansari, Aisha and
Ashraf, Ahmed and
Alturayeif, Nora and
Alcoba Inciarte, Alcides and
Elmadany, AbdelRahim A. and
Tourad, Mohamedou Cheikh and
Berrada, Ismail and
Jarrar, Mustafa and
Shehata, Shady and
Abdul-Mageed, Muhammad",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.1254/",
pages = "23048--23079",
ISBN = "979-8-89176-335-7",
abstract = "Mainstream large vision-language models (LVLMs) inherently encode cultural biases, highlighting the need for diverse multimodal datasets. To address this gap, we introduce PEARL, a large-scale Arabic multimodal dataset and benchmark explicitly designed for cultural understanding. Constructed through advanced agentic workflows and extensive human-in-the-loop annotations by 37 annotators from across the Arab world, PEARL comprises over 309K multimodal examples spanning ten culturally significant domains covering all Arab countries. We further provide two robust evaluation benchmarks (PEARL and PEARL-LITE) along with a specialized subset (PEARL-X) explicitly developed to assess nuanced cultural variations. Comprehensive evaluations on state-of-the-art open and proprietary LVLMs demonstrate that reasoning-centric instruction alignment substantially improves models' cultural grounding compared to conventional scaling methods. PEARL establishes a foundational resource for advancing culturally-informed multimodal modeling research. All datasets and benchmarks are publicly available."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="alwajih-etal-2025-pearl">
<titleInfo>
<title>Pearl: A Multimodal Culturally-Aware Arabic Instruction Dataset</title>
</titleInfo>
<name type="personal">
<namePart type="given">Fakhraddin</namePart>
<namePart type="family">Alwajih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samar</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Magdy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdellah</namePart>
<namePart type="family">El Mekki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Omer</namePart>
<namePart type="family">Nacar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Youssef</namePart>
<namePart type="family">Nafea</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Safaa</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Abdelfadil</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdulfattah</namePart>
<namePart type="given">Mohammed</namePart>
<namePart type="family">Yahya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hamzah</namePart>
<namePart type="family">Luqman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nada</namePart>
<namePart type="family">Almarwani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samah</namePart>
<namePart type="family">Aloufi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Baraah</namePart>
<namePart type="family">Qawasmeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Houdaifa</namePart>
<namePart type="family">Atou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Serry</namePart>
<namePart type="family">Sibaee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hamzah</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Alsayadi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Walid</namePart>
<namePart type="family">Al-Dhabyani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maged</namePart>
<namePart type="given">S</namePart>
<namePart type="family">Al-shaibani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aya</namePart>
<namePart type="family">El aatar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nour</namePart>
<namePart type="family">Qandos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rahaf</namePart>
<namePart type="family">Alhamouri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samar</namePart>
<namePart type="family">Ahmad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammed</namePart>
<namePart type="given">Anwar</namePart>
<namePart type="family">AL-Ghrawi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aminetou</namePart>
<namePart type="family">Yacoub</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruwa</namePart>
<namePart type="family">AbuHweidi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vatimetou</namePart>
<namePart type="given">Mohamed</namePart>
<namePart type="family">Lemin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Reem</namePart>
<namePart type="family">Abdel-Salam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahlam</namePart>
<namePart type="family">Bashiti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adel</namePart>
<namePart type="family">Ammar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aisha</namePart>
<namePart type="family">Alansari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="family">Ashraf</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nora</namePart>
<namePart type="family">Alturayeif</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alcides</namePart>
<namePart type="family">Alcoba Inciarte</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">AbdelRahim</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Elmadany</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamedou</namePart>
<namePart type="given">Cheikh</namePart>
<namePart type="family">Tourad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ismail</namePart>
<namePart type="family">Berrada</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mustafa</namePart>
<namePart type="family">Jarrar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shady</namePart>
<namePart type="family">Shehata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="family">Abdul-Mageed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>Mainstream large vision-language models (LVLMs) inherently encode cultural biases, highlighting the need for diverse multimodal datasets. To address this gap, we introduce PEARL, a large-scale Arabic multimodal dataset and benchmark explicitly designed for cultural understanding. Constructed through advanced agentic workflows and extensive human-in-the-loop annotations by 37 annotators from across the Arab world, PEARL comprises over 309K multimodal examples spanning ten culturally significant domains covering all Arab countries. We further provide two robust evaluation benchmarks (PEARL and PEARL-LITE) along with a specialized subset (PEARL-X) explicitly developed to assess nuanced cultural variations. Comprehensive evaluations on state-of-the-art open and proprietary LVLMs demonstrate that reasoning-centric instruction alignment substantially improves models’ cultural grounding compared to conventional scaling methods. PEARL establishes a foundational resource for advancing culturally-informed multimodal modeling research. All datasets and benchmarks are publicly available.</abstract>
<identifier type="citekey">alwajih-etal-2025-pearl</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.1254/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>23048</start>
<end>23079</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Pearl: A Multimodal Culturally-Aware Arabic Instruction Dataset
%A Alwajih, Fakhraddin
%A Magdy, Samar M.
%A El Mekki, Abdellah
%A Nacar, Omer
%A Nafea, Youssef
%A Abdelfadil, Safaa Taher
%A Yahya, Abdulfattah Mohammed
%A Luqman, Hamzah
%A Almarwani, Nada
%A Aloufi, Samah
%A Qawasmeh, Baraah
%A Atou, Houdaifa
%A Sibaee, Serry
%A Alsayadi, Hamzah A.
%A Al-Dhabyani, Walid
%A Al-shaibani, Maged S.
%A El aatar, Aya
%A Qandos, Nour
%A Alhamouri, Rahaf
%A Ahmad, Samar
%A AL-Ghrawi, Mohammed Anwar
%A Yacoub, Aminetou
%A AbuHweidi, Ruwa
%A Lemin, Vatimetou Mohamed
%A Abdel-Salam, Reem
%A Bashiti, Ahlam
%A Ammar, Adel
%A Alansari, Aisha
%A Ashraf, Ahmed
%A Alturayeif, Nora
%A Alcoba Inciarte, Alcides
%A Elmadany, AbdelRahim A.
%A Tourad, Mohamedou Cheikh
%A Berrada, Ismail
%A Jarrar, Mustafa
%A Shehata, Shady
%A Abdul-Mageed, Muhammad
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F alwajih-etal-2025-pearl
%X Mainstream large vision-language models (LVLMs) inherently encode cultural biases, highlighting the need for diverse multimodal datasets. To address this gap, we introduce PEARL, a large-scale Arabic multimodal dataset and benchmark explicitly designed for cultural understanding. Constructed through advanced agentic workflows and extensive human-in-the-loop annotations by 37 annotators from across the Arab world, PEARL comprises over 309K multimodal examples spanning ten culturally significant domains covering all Arab countries. We further provide two robust evaluation benchmarks (PEARL and PEARL-LITE) along with a specialized subset (PEARL-X) explicitly developed to assess nuanced cultural variations. Comprehensive evaluations on state-of-the-art open and proprietary LVLMs demonstrate that reasoning-centric instruction alignment substantially improves models’ cultural grounding compared to conventional scaling methods. PEARL establishes a foundational resource for advancing culturally-informed multimodal modeling research. All datasets and benchmarks are publicly available.
%U https://aclanthology.org/2025.findings-emnlp.1254/
%P 23048-23079
Markdown (Informal)
[Pearl: A Multimodal Culturally-Aware Arabic Instruction Dataset](https://aclanthology.org/2025.findings-emnlp.1254/) (Alwajih et al., Findings 2025)
ACL
- Fakhraddin Alwajih, Samar M. Magdy, Abdellah El Mekki, Omer Nacar, Youssef Nafea, Safaa Taher Abdelfadil, Abdulfattah Mohammed Yahya, Hamzah Luqman, Nada Almarwani, Samah Aloufi, Baraah Qawasmeh, Houdaifa Atou, Serry Sibaee, Hamzah A. Alsayadi, Walid Al-Dhabyani, Maged S. Al-shaibani, Aya El aatar, Nour Qandos, Rahaf Alhamouri, Samar Ahmad, Mohammed Anwar AL-Ghrawi, Aminetou Yacoub, Ruwa AbuHweidi, Vatimetou Mohamed Lemin, Reem Abdel-Salam, Ahlam Bashiti, Adel Ammar, Aisha Alansari, Ahmed Ashraf, Nora Alturayeif, Alcides Alcoba Inciarte, AbdelRahim A. Elmadany, Mohamedou Cheikh Tourad, Ismail Berrada, Mustafa Jarrar, Shady Shehata, and Muhammad Abdul-Mageed. 2025. Pearl: A Multimodal Culturally-Aware Arabic Instruction Dataset. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 23048–23079, Suzhou, China. Association for Computational Linguistics.