@inproceedings{ghaboura-etal-2026-arb,
title = "{ARB}: A Comprehensive {A}rabic Multimodal Reasoning Benchmark",
author = "Ghaboura, Sara and
Patle, Shubham and
More, Ketan and
Alghallabi, Wafa Hamad Mohamed and
Thawakar, Omkar and
Laaksonen, Jorma and
Cholakkal, Hisham and
Khan, Salman and
Anwer, Rao",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.723/",
doi = "10.63317/2gyndds6s2us",
pages = "9202--9216",
abstract = "As Large Multimodal Models (LMMs) become more capable, there is growing interest in evaluating their reasoning processes alongside their final outputs. However, most existing benchmarks remain focused on English, overlooking languages with rich linguistic and cultural depth such as Arabic. To address this gap, we introduce the Comprehensive Arabic Multimodal Reasoning Benchmark (ARB), the first benchmark designed to evaluate step-by-step reasoning in Arabic across both textual and visual modalities. ARB covers 11 diverse domains and over 40 subfields, including visual reasoning, optical character recognition, scientific analysis, and cultural interpretation. It comprises 2,219 multimodal samples paired with over 8K human-curated reasoning steps and corresponding actions, verified through a human-in-the-loop process. We evaluated 15 state-of-the-art open- and closed-source LMMs and found persistent challenges in coherence, faithfulness, and cultural grounding. ARB provides a structured framework for diagnosing multimodal reasoning in underrepresented languages, marking a critical step toward inclusive, transparent, and culturally aware AI systems. The benchmark, rubric, and evaluation suite are publicly available"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ghaboura-etal-2026-arb">
<titleInfo>
<title>ARB: A Comprehensive Arabic Multimodal Reasoning Benchmark</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sara</namePart>
<namePart type="family">Ghaboura</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shubham</namePart>
<namePart type="family">Patle</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ketan</namePart>
<namePart type="family">More</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wafa</namePart>
<namePart type="given">Hamad</namePart>
<namePart type="given">Mohamed</namePart>
<namePart type="family">Alghallabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Omkar</namePart>
<namePart type="family">Thawakar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jorma</namePart>
<namePart type="family">Laaksonen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hisham</namePart>
<namePart type="family">Cholakkal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salman</namePart>
<namePart type="family">Khan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rao</namePart>
<namePart type="family">Anwer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>As Large Multimodal Models (LMMs) become more capable, there is growing interest in evaluating their reasoning processes alongside their final outputs. However, most existing benchmarks remain focused on English, overlooking languages with rich linguistic and cultural depth such as Arabic. To address this gap, we introduce the Comprehensive Arabic Multimodal Reasoning Benchmark (ARB), the first benchmark designed to evaluate step-by-step reasoning in Arabic across both textual and visual modalities. ARB covers 11 diverse domains and over 40 subfields, including visual reasoning, optical character recognition, scientific analysis, and cultural interpretation. It comprises 2,219 multimodal samples paired with over 8K human-curated reasoning steps and corresponding actions, verified through a human-in-the-loop process. We evaluated 15 state-of-the-art open- and closed-source LMMs and found persistent challenges in coherence, faithfulness, and cultural grounding. ARB provides a structured framework for diagnosing multimodal reasoning in underrepresented languages, marking a critical step toward inclusive, transparent, and culturally aware AI systems. The benchmark, rubric, and evaluation suite are publicly available</abstract>
<identifier type="citekey">ghaboura-etal-2026-arb</identifier>
<identifier type="doi">10.63317/2gyndds6s2us</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.723/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>9202</start>
<end>9216</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ARB: A Comprehensive Arabic Multimodal Reasoning Benchmark
%A Ghaboura, Sara
%A Patle, Shubham
%A More, Ketan
%A Alghallabi, Wafa Hamad Mohamed
%A Thawakar, Omkar
%A Laaksonen, Jorma
%A Cholakkal, Hisham
%A Khan, Salman
%A Anwer, Rao
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F ghaboura-etal-2026-arb
%X As Large Multimodal Models (LMMs) become more capable, there is growing interest in evaluating their reasoning processes alongside their final outputs. However, most existing benchmarks remain focused on English, overlooking languages with rich linguistic and cultural depth such as Arabic. To address this gap, we introduce the Comprehensive Arabic Multimodal Reasoning Benchmark (ARB), the first benchmark designed to evaluate step-by-step reasoning in Arabic across both textual and visual modalities. ARB covers 11 diverse domains and over 40 subfields, including visual reasoning, optical character recognition, scientific analysis, and cultural interpretation. It comprises 2,219 multimodal samples paired with over 8K human-curated reasoning steps and corresponding actions, verified through a human-in-the-loop process. We evaluated 15 state-of-the-art open- and closed-source LMMs and found persistent challenges in coherence, faithfulness, and cultural grounding. ARB provides a structured framework for diagnosing multimodal reasoning in underrepresented languages, marking a critical step toward inclusive, transparent, and culturally aware AI systems. The benchmark, rubric, and evaluation suite are publicly available
%R 10.63317/2gyndds6s2us
%U https://aclanthology.org/2026.lrec-1.723/
%U https://doi.org/10.63317/2gyndds6s2us
%P 9202-9216
Markdown (Informal)
[ARB: A Comprehensive Arabic Multimodal Reasoning Benchmark](https://aclanthology.org/2026.lrec-1.723/) (Ghaboura et al., LREC 2026)
ACL
- Sara Ghaboura, Shubham Patle, Ketan More, Wafa Hamad Mohamed Alghallabi, Omkar Thawakar, Jorma Laaksonen, Hisham Cholakkal, Salman Khan, and Rao Anwer. 2026. ARB: A Comprehensive Arabic Multimodal Reasoning Benchmark. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 9202–9216, Palma de Mallorca, Spain. ELRA Language Resource Association.