@inproceedings{semon-islam-2026-cuet,
title = "{CUET}{\_}{I}nfer{X} at {I}mage{E}val 2026 Shared Tasks: Audio-Visual {L}o{RA} Adaptation of {Q}wen2.5-Omni for Culturally-Grounded Spoken {VQA}",
author = "Semon, Md. Ashraful Islam and
Islam, Jihadul",
editor = "Abdul-Mageed, Muhammad and
Elsayed, Tamer and
Bouamor, Houda and
El-Beltagy, Samhaa and
Al-Khalifa, Hend and
Abdelali, Ahmed and
Elmadany, AbdelRahim and
El Mekki, Abdellah and
Darwish, Kareem and
Alkhereyf, Sakhar and
Luqman, Hamzah and
Khalifa, Salam and
Magdy, Samar and
Alyafeai, Zaid and
Hamed, Injy and
Zaghouani, Wajdi and
Zitouni, Imed",
booktitle = "Proceedings of The Fourth {A}rabic Natural Language Processing Conference: Shared Tasks",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.arabicnlp-sharedtasks.93/",
pages = "747--752",
ISBN = "979-8-89176-497-2",
abstract = "We present CUET{\_}InferX, our submission to Subtask{~}1a (Spoken VQA, English-audio) of the ImageEval 2026 shared task on the AynVQA dataset. The task requires predicting the correct answer option solely from an image and a raw audio clip containing both a spoken question and three candidate options. We fine-tune \texttt{Qwen2.5-Omni-3B} with LoRA applied exclusively to the language reasoning backbone, keeping both audio and vision encoders frozen, under a single 16 GB Kaggle T4 GPU. Training uses a time-budgeted, class-stratified subsample of 2,055 examples. Our system achieves \textbf{91.20{\%}} test accuracy (Balanced Accuracy 87.72{\%}, Macro-F1 80.50{\%}), placing 3rd position, with gains of $+7.20$ accuracy and $+15.95$ Macro-F1 over the zero-shot base. We compare against a cascaded ASR$\to$VLM alternative and zero-shot 3B/7B baselines, and explain with devtest evidence why we submitted the end-to-end fine-tuned model despite lower raw test accuracy than the cascade."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="semon-islam-2026-cuet">
<titleInfo>
<title>CUET_InferX at ImageEval 2026 Shared Tasks: Audio-Visual LoRA Adaptation of Qwen2.5-Omni for Culturally-Grounded Spoken VQA</title>
</titleInfo>
<name type="personal">
<namePart type="given">Md.</namePart>
<namePart type="given">Ashraful</namePart>
<namePart type="given">Islam</namePart>
<namePart type="family">Semon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jihadul</namePart>
<namePart type="family">Islam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="family">Abdul-Mageed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tamer</namePart>
<namePart type="family">Elsayed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samhaa</namePart>
<namePart type="family">El-Beltagy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hend</namePart>
<namePart type="family">Al-Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="family">Abdelali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">AbdelRahim</namePart>
<namePart type="family">Elmadany</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdellah</namePart>
<namePart type="family">El Mekki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kareem</namePart>
<namePart type="family">Darwish</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakhar</namePart>
<namePart type="family">Alkhereyf</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hamzah</namePart>
<namePart type="family">Luqman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salam</namePart>
<namePart type="family">Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samar</namePart>
<namePart type="family">Magdy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zaid</namePart>
<namePart type="family">Alyafeai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Injy</namePart>
<namePart type="family">Hamed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wajdi</namePart>
<namePart type="family">Zaghouani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Imed</namePart>
<namePart type="family">Zitouni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-497-2</identifier>
</relatedItem>
<abstract>We present CUET_InferX, our submission to Subtask 1a (Spoken VQA, English-audio) of the ImageEval 2026 shared task on the AynVQA dataset. The task requires predicting the correct answer option solely from an image and a raw audio clip containing both a spoken question and three candidate options. We fine-tune Qwen2.5-Omni-3B with LoRA applied exclusively to the language reasoning backbone, keeping both audio and vision encoders frozen, under a single 16 GB Kaggle T4 GPU. Training uses a time-budgeted, class-stratified subsample of 2,055 examples. Our system achieves 91.20% test accuracy (Balanced Accuracy 87.72%, Macro-F1 80.50%), placing 3rd position, with gains of +7.20 accuracy and +15.95 Macro-F1 over the zero-shot base. We compare against a cascaded ASRVLM alternative and zero-shot 3B/7B baselines, and explain with devtest evidence why we submitted the end-to-end fine-tuned model despite lower raw test accuracy than the cascade.</abstract>
<identifier type="citekey">semon-islam-2026-cuet</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.93/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>747</start>
<end>752</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CUET_InferX at ImageEval 2026 Shared Tasks: Audio-Visual LoRA Adaptation of Qwen2.5-Omni for Culturally-Grounded Spoken VQA
%A Semon, Md. Ashraful Islam
%A Islam, Jihadul
%Y Abdul-Mageed, Muhammad
%Y Elsayed, Tamer
%Y Bouamor, Houda
%Y El-Beltagy, Samhaa
%Y Al-Khalifa, Hend
%Y Abdelali, Ahmed
%Y Elmadany, AbdelRahim
%Y El Mekki, Abdellah
%Y Darwish, Kareem
%Y Alkhereyf, Sakhar
%Y Luqman, Hamzah
%Y Khalifa, Salam
%Y Magdy, Samar
%Y Alyafeai, Zaid
%Y Hamed, Injy
%Y Zaghouani, Wajdi
%Y Zitouni, Imed
%S Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-497-2
%F semon-islam-2026-cuet
%X We present CUET_InferX, our submission to Subtask 1a (Spoken VQA, English-audio) of the ImageEval 2026 shared task on the AynVQA dataset. The task requires predicting the correct answer option solely from an image and a raw audio clip containing both a spoken question and three candidate options. We fine-tune Qwen2.5-Omni-3B with LoRA applied exclusively to the language reasoning backbone, keeping both audio and vision encoders frozen, under a single 16 GB Kaggle T4 GPU. Training uses a time-budgeted, class-stratified subsample of 2,055 examples. Our system achieves 91.20% test accuracy (Balanced Accuracy 87.72%, Macro-F1 80.50%), placing 3rd position, with gains of +7.20 accuracy and +15.95 Macro-F1 over the zero-shot base. We compare against a cascaded ASRVLM alternative and zero-shot 3B/7B baselines, and explain with devtest evidence why we submitted the end-to-end fine-tuned model despite lower raw test accuracy than the cascade.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.93/
%P 747-752
Markdown (Informal)
[CUET_InferX at ImageEval 2026 Shared Tasks: Audio-Visual LoRA Adaptation of Qwen2.5-Omni for Culturally-Grounded Spoken VQA](https://aclanthology.org/2026.arabicnlp-sharedtasks.93/) (Semon & Islam, ArabicNLP 2026)
ACL