@inproceedings{wei-etal-2026-multimodal,
title = "Do Multimodal {LLM}s Understand Order? Measuring the Fragility of Multimodal Reasoning under Input Order Perturbations",
author = "Wei, Sheng-Lun and
Liao, Yu-Ling and
Huang, Hen-Hsen and
Chen, Hsin-Hsi",
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-1.716/",
doi = "10.63317/4jtpgzks8pbr",
pages = "9118--9128",
abstract = "Multimodal reasoning has progressed rapidly with large vision-language models (LVLMs), yet their robustness under input variations remains underexplored. This study investigates positional bias in LVLMs for multimodal multiple-choice questions. Our analysis shows that model predictions are sensitive to both choice and modality ordering. We conduct a large-scale evaluation on MMMU, CVQA, and MMBench using fourteen representative models. Further analysis examines how question properties, including difficulty, domain, and image type, affect robustness. We also assess whether text-based mitigation strategies transfer to the VQA setting and perform ablation studies on self-consistency and reasoning complexity. Overall, our findings provide the first comprehensive understanding of positional bias from a vision-language perspective, highlighting key challenges in achieving stable multimodal reasoning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wei-etal-2026-multimodal">
<titleInfo>
<title>Do Multimodal LLMs Understand Order? Measuring the Fragility of Multimodal Reasoning under Input Order Perturbations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sheng-Lun</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu-Ling</namePart>
<namePart type="family">Liao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hen-Hsen</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hsin-Hsi</namePart>
<namePart type="family">Chen</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>Multimodal reasoning has progressed rapidly with large vision-language models (LVLMs), yet their robustness under input variations remains underexplored. This study investigates positional bias in LVLMs for multimodal multiple-choice questions. Our analysis shows that model predictions are sensitive to both choice and modality ordering. We conduct a large-scale evaluation on MMMU, CVQA, and MMBench using fourteen representative models. Further analysis examines how question properties, including difficulty, domain, and image type, affect robustness. We also assess whether text-based mitigation strategies transfer to the VQA setting and perform ablation studies on self-consistency and reasoning complexity. Overall, our findings provide the first comprehensive understanding of positional bias from a vision-language perspective, highlighting key challenges in achieving stable multimodal reasoning.</abstract>
<identifier type="citekey">wei-etal-2026-multimodal</identifier>
<identifier type="doi">10.63317/4jtpgzks8pbr</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.716/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>9118</start>
<end>9128</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Do Multimodal LLMs Understand Order? Measuring the Fragility of Multimodal Reasoning under Input Order Perturbations
%A Wei, Sheng-Lun
%A Liao, Yu-Ling
%A Huang, Hen-Hsen
%A Chen, Hsin-Hsi
%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 wei-etal-2026-multimodal
%X Multimodal reasoning has progressed rapidly with large vision-language models (LVLMs), yet their robustness under input variations remains underexplored. This study investigates positional bias in LVLMs for multimodal multiple-choice questions. Our analysis shows that model predictions are sensitive to both choice and modality ordering. We conduct a large-scale evaluation on MMMU, CVQA, and MMBench using fourteen representative models. Further analysis examines how question properties, including difficulty, domain, and image type, affect robustness. We also assess whether text-based mitigation strategies transfer to the VQA setting and perform ablation studies on self-consistency and reasoning complexity. Overall, our findings provide the first comprehensive understanding of positional bias from a vision-language perspective, highlighting key challenges in achieving stable multimodal reasoning.
%R 10.63317/4jtpgzks8pbr
%U https://aclanthology.org/2026.lrec-1.716/
%U https://doi.org/10.63317/4jtpgzks8pbr
%P 9118-9128
Markdown (Informal)
[Do Multimodal LLMs Understand Order? Measuring the Fragility of Multimodal Reasoning under Input Order Perturbations](https://aclanthology.org/2026.lrec-1.716/) (Wei et al., LREC 2026)
ACL