@inproceedings{tonny-etal-2026-quadcore,
title = "{Q}uad{C}ore at {A}r{G}uard Shared Tasks: A Simple Threshold Calibration Approach for {A}rabic Hateful Meme Detection",
author = "Tonny, Umme Saima and
Tabassum, Maimuna and
Muntaha, Sidratul and
Anzum, Sabila",
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.64/",
pages = "490--496",
ISBN = "979-8-89176-497-2",
abstract = "We describe our system for Subtask A1 (binary hateful meme detection) of the ArGuard 2026 shared task (Alam et al., 2026). Our final system fine-tunes MARBERTv2 with class-weighted loss and macro-F1-based checkpoint selection, then applies a post-hoc decision-threshold calibration to correct for a class-distribution shift we found between the training/development splits (37.8{\%} Hateful) and the held-out test split (29.6{\%} Hateful). This calibration step requires no retraining and adds 3.3 macro-F1 points (0.7286 {\textrightarrow} 0.7616) on the official test split. On the same test split, this calibrated system also outperforms two alternatives we evaluated there {--} multi-seed ensembling and naive multimodal late-fusion. They match or exceed backbone-substitution and multi-model-ensemble variants that we were only able to evaluate on local development data. We report our full experimental log, including the negative results, with each result explicitly labeled by evaluation split, and compare our score to the AHA-Memes benchmark, where a much larger vision-language model reaches a similar range. Code and configuration files are released for reproducibility."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tonny-etal-2026-quadcore">
<titleInfo>
<title>QuadCore at ArGuard Shared Tasks: A Simple Threshold Calibration Approach for Arabic Hateful Meme Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Umme</namePart>
<namePart type="given">Saima</namePart>
<namePart type="family">Tonny</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maimuna</namePart>
<namePart type="family">Tabassum</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sidratul</namePart>
<namePart type="family">Muntaha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sabila</namePart>
<namePart type="family">Anzum</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 describe our system for Subtask A1 (binary hateful meme detection) of the ArGuard 2026 shared task (Alam et al., 2026). Our final system fine-tunes MARBERTv2 with class-weighted loss and macro-F1-based checkpoint selection, then applies a post-hoc decision-threshold calibration to correct for a class-distribution shift we found between the training/development splits (37.8% Hateful) and the held-out test split (29.6% Hateful). This calibration step requires no retraining and adds 3.3 macro-F1 points (0.7286 → 0.7616) on the official test split. On the same test split, this calibrated system also outperforms two alternatives we evaluated there – multi-seed ensembling and naive multimodal late-fusion. They match or exceed backbone-substitution and multi-model-ensemble variants that we were only able to evaluate on local development data. We report our full experimental log, including the negative results, with each result explicitly labeled by evaluation split, and compare our score to the AHA-Memes benchmark, where a much larger vision-language model reaches a similar range. Code and configuration files are released for reproducibility.</abstract>
<identifier type="citekey">tonny-etal-2026-quadcore</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.64/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>490</start>
<end>496</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T QuadCore at ArGuard Shared Tasks: A Simple Threshold Calibration Approach for Arabic Hateful Meme Detection
%A Tonny, Umme Saima
%A Tabassum, Maimuna
%A Muntaha, Sidratul
%A Anzum, Sabila
%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 tonny-etal-2026-quadcore
%X We describe our system for Subtask A1 (binary hateful meme detection) of the ArGuard 2026 shared task (Alam et al., 2026). Our final system fine-tunes MARBERTv2 with class-weighted loss and macro-F1-based checkpoint selection, then applies a post-hoc decision-threshold calibration to correct for a class-distribution shift we found between the training/development splits (37.8% Hateful) and the held-out test split (29.6% Hateful). This calibration step requires no retraining and adds 3.3 macro-F1 points (0.7286 → 0.7616) on the official test split. On the same test split, this calibrated system also outperforms two alternatives we evaluated there – multi-seed ensembling and naive multimodal late-fusion. They match or exceed backbone-substitution and multi-model-ensemble variants that we were only able to evaluate on local development data. We report our full experimental log, including the negative results, with each result explicitly labeled by evaluation split, and compare our score to the AHA-Memes benchmark, where a much larger vision-language model reaches a similar range. Code and configuration files are released for reproducibility.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.64/
%P 490-496
Markdown (Informal)
[QuadCore at ArGuard Shared Tasks: A Simple Threshold Calibration Approach for Arabic Hateful Meme Detection](https://aclanthology.org/2026.arabicnlp-sharedtasks.64/) (Tonny et al., ArabicNLP 2026)
ACL