@inproceedings{zaher-etal-2026-zhr,
title = "{ZHR} at {S}tance{E}val-2026: Surface-Matched Synthetic Demonstrations and Metric-Aware Decoding for {A}rabic Cross-Target Stance Detection",
author = "Zaher, Mahmoud and
Bahgat, Mohamed and
Alshrief, Mahmoud",
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.83/",
pages = "656--666",
ISBN = "979-8-89176-497-2",
abstract = "We describe the ZHR system for both tracks of StanceEval-2026, Arabic stance detection on Mawqif-XT, where Track 1 evaluates on seen targets and Track 2 on unseen ones. The system combines fine-tuned Arabic encoders, a few-shot classifier that retrieves labelled examples for a served instruction model, and generative models adapted with LoRA, and turns their blended probabilities into labels with a threshold rule that abstains to None unless one stance class is probable enough, with thresholds fixed on the development split. We report the whole sequence of experiments, including eight that lowered $F_{\mathrm{avg}2}$ and were not adopted. The main finding is that transfer to a new target is carried by the demonstrations, the labelled examples retrieved into the prompt, rather than by the model itself. Generated on-target tweets in the example pool gave the largest gain on both tracks, matching the surface form of the test tweets helped where the target has a distinctive style, and doubling the examples hurt. Our official results are $F_{\mathrm{avg}2}$ 0.896 with accuracy 0.855 on Track 1, and $F_{\mathrm{avg}2}$ 0.940 with accuracy 0.922 on Track 2. The strongest official baselines reach 0.8425 on the Track 1 development set and 0.7475 on the Track 2 test set."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zaher-etal-2026-zhr">
<titleInfo>
<title>ZHR at StanceEval-2026: Surface-Matched Synthetic Demonstrations and Metric-Aware Decoding for Arabic Cross-Target Stance Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mahmoud</namePart>
<namePart type="family">Zaher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="family">Bahgat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mahmoud</namePart>
<namePart type="family">Alshrief</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 the ZHR system for both tracks of StanceEval-2026, Arabic stance detection on Mawqif-XT, where Track 1 evaluates on seen targets and Track 2 on unseen ones. The system combines fine-tuned Arabic encoders, a few-shot classifier that retrieves labelled examples for a served instruction model, and generative models adapted with LoRA, and turns their blended probabilities into labels with a threshold rule that abstains to None unless one stance class is probable enough, with thresholds fixed on the development split. We report the whole sequence of experiments, including eight that lowered F_avg2 and were not adopted. The main finding is that transfer to a new target is carried by the demonstrations, the labelled examples retrieved into the prompt, rather than by the model itself. Generated on-target tweets in the example pool gave the largest gain on both tracks, matching the surface form of the test tweets helped where the target has a distinctive style, and doubling the examples hurt. Our official results are F_avg2 0.896 with accuracy 0.855 on Track 1, and F_avg2 0.940 with accuracy 0.922 on Track 2. The strongest official baselines reach 0.8425 on the Track 1 development set and 0.7475 on the Track 2 test set.</abstract>
<identifier type="citekey">zaher-etal-2026-zhr</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.83/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>656</start>
<end>666</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ZHR at StanceEval-2026: Surface-Matched Synthetic Demonstrations and Metric-Aware Decoding for Arabic Cross-Target Stance Detection
%A Zaher, Mahmoud
%A Bahgat, Mohamed
%A Alshrief, Mahmoud
%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 zaher-etal-2026-zhr
%X We describe the ZHR system for both tracks of StanceEval-2026, Arabic stance detection on Mawqif-XT, where Track 1 evaluates on seen targets and Track 2 on unseen ones. The system combines fine-tuned Arabic encoders, a few-shot classifier that retrieves labelled examples for a served instruction model, and generative models adapted with LoRA, and turns their blended probabilities into labels with a threshold rule that abstains to None unless one stance class is probable enough, with thresholds fixed on the development split. We report the whole sequence of experiments, including eight that lowered F_avg2 and were not adopted. The main finding is that transfer to a new target is carried by the demonstrations, the labelled examples retrieved into the prompt, rather than by the model itself. Generated on-target tweets in the example pool gave the largest gain on both tracks, matching the surface form of the test tweets helped where the target has a distinctive style, and doubling the examples hurt. Our official results are F_avg2 0.896 with accuracy 0.855 on Track 1, and F_avg2 0.940 with accuracy 0.922 on Track 2. The strongest official baselines reach 0.8425 on the Track 1 development set and 0.7475 on the Track 2 test set.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.83/
%P 656-666
Markdown (Informal)
[ZHR at StanceEval-2026: Surface-Matched Synthetic Demonstrations and Metric-Aware Decoding for Arabic Cross-Target Stance Detection](https://aclanthology.org/2026.arabicnlp-sharedtasks.83/) (Zaher et al., ArabicNLP 2026)
ACL