@inproceedings{dang-etal-2026-vmmu,
title = "{VMMU}: A {V}ietnamese Multitask Multimodal Understanding and Reasoning Benchmark",
author = "Dang, Vy Tuong and
Vo, An and
Villa-Cueva, Emilio and
Tau, Quang and
Dm, Duc and
Solorio, Thamar and
Kim, Daeyoung",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {IJCNLP}-{AACL} 2026",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-aacl.44/",
pages = "825--870",
ISBN = "979-8-89176-444-6",
abstract = "Vision language models are predominantly evaluated on English-centric multimodal benchmarks, leaving their behavior in other languages and cultures insufficiently understood. We introduce VMMU, a Vietnamese Multitask Multimodal Understanding and Reasoning benchmark of 2,548 multiple-choice questions across seven domains: Mathematics, Physics, Chemistry, Biology, Geography, Driving Test, and IQ Test. Every question requires jointly reasoning over Vietnamese text and non-text visual evidence (charts, diagrams, tables, traffic scenes), and is distributed as a single rendered page image. We evaluate 10 open-source and 5 closed-source VLMs on VMMU. Test-time compute improves performance from 57{\%} to 78{\%} accuracy, but it remains inferior to an expert baseline (99{\%} accuracy). Extensive error analyses show that closed-source VLMs already achieve strong Vietnamese OCR performance, yet still struggle on VMMU. This suggests that the primary bottleneck is multimodal grounding and reasoning rather than text recognition or Vietnamese language understanding. Code and data are available at: https://vmmu-bench.github.io."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dang-etal-2026-vmmu">
<titleInfo>
<title>VMMU: A Vietnamese Multitask Multimodal Understanding and Reasoning Benchmark</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vy</namePart>
<namePart type="given">Tuong</namePart>
<namePart type="family">Dang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">An</namePart>
<namePart type="family">Vo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Emilio</namePart>
<namePart type="family">Villa-Cueva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Quang</namePart>
<namePart type="family">Tau</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Duc</namePart>
<namePart type="family">Dm</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thamar</namePart>
<namePart type="family">Solorio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daeyoung</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-444-6</identifier>
</relatedItem>
<abstract>Vision language models are predominantly evaluated on English-centric multimodal benchmarks, leaving their behavior in other languages and cultures insufficiently understood. We introduce VMMU, a Vietnamese Multitask Multimodal Understanding and Reasoning benchmark of 2,548 multiple-choice questions across seven domains: Mathematics, Physics, Chemistry, Biology, Geography, Driving Test, and IQ Test. Every question requires jointly reasoning over Vietnamese text and non-text visual evidence (charts, diagrams, tables, traffic scenes), and is distributed as a single rendered page image. We evaluate 10 open-source and 5 closed-source VLMs on VMMU. Test-time compute improves performance from 57% to 78% accuracy, but it remains inferior to an expert baseline (99% accuracy). Extensive error analyses show that closed-source VLMs already achieve strong Vietnamese OCR performance, yet still struggle on VMMU. This suggests that the primary bottleneck is multimodal grounding and reasoning rather than text recognition or Vietnamese language understanding. Code and data are available at: https://vmmu-bench.github.io.</abstract>
<identifier type="citekey">dang-etal-2026-vmmu</identifier>
<location>
<url>https://aclanthology.org/2026.findings-aacl.44/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>825</start>
<end>870</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T VMMU: A Vietnamese Multitask Multimodal Understanding and Reasoning Benchmark
%A Dang, Vy Tuong
%A Vo, An
%A Villa-Cueva, Emilio
%A Tau, Quang
%A Dm, Duc
%A Solorio, Thamar
%A Kim, Daeyoung
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F dang-etal-2026-vmmu
%X Vision language models are predominantly evaluated on English-centric multimodal benchmarks, leaving their behavior in other languages and cultures insufficiently understood. We introduce VMMU, a Vietnamese Multitask Multimodal Understanding and Reasoning benchmark of 2,548 multiple-choice questions across seven domains: Mathematics, Physics, Chemistry, Biology, Geography, Driving Test, and IQ Test. Every question requires jointly reasoning over Vietnamese text and non-text visual evidence (charts, diagrams, tables, traffic scenes), and is distributed as a single rendered page image. We evaluate 10 open-source and 5 closed-source VLMs on VMMU. Test-time compute improves performance from 57% to 78% accuracy, but it remains inferior to an expert baseline (99% accuracy). Extensive error analyses show that closed-source VLMs already achieve strong Vietnamese OCR performance, yet still struggle on VMMU. This suggests that the primary bottleneck is multimodal grounding and reasoning rather than text recognition or Vietnamese language understanding. Code and data are available at: https://vmmu-bench.github.io.
%U https://aclanthology.org/2026.findings-aacl.44/
%P 825-870
Markdown (Informal)
[VMMU: A Vietnamese Multitask Multimodal Understanding and Reasoning Benchmark](https://aclanthology.org/2026.findings-aacl.44/) (Dang et al., Findings 2026)
ACL