@inproceedings{alshehri-etal-2026-durham-jazan,
title = "{D}urham-Jazan at {AISA}-{A}rabic{FC}: A Field-Wise {QL}o{RA} Ensemble and the Sensitivity of Exact-Match Scoring",
author = "Alshehri, Amal Saad and
Bencomo, Nelly and
Babiker, Housam and
Atapour-Abarghouei, Amir",
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.200/",
pages = "1552--1563",
ISBN = "979-8-89176-497-2",
abstract = "We describe the Durham-Jazan submission to AISA-ArabicFC 2026, an Arabic function-calling task in which a system selects one tool from four candidates and fills its arguments, scored on tool choice (FnAcc) and on strict exact match of the whole argument dictionary (ArgEM). Across six QLoRA fine-tunes of ALLaM-7B and Qwen2.5-7B, trained on the provided data, negatives derived from it and 68 hand-written no-tool utterances, and run on one 16 GB GPU, we vote on each argument separately, then apply two deterministic rules estimated from training statistics. The system reaches ArgEM 0.824 on development under the final evaluator and 0.826 on the blind test, placing 9th of 17 in Track A, 8th of 17 in Track B and 11th of 21 in Track C. Our main findings concern the benchmark. The reference tool is listed first on every training and development row that calls one, in every data release. Shuffling the candidates at inference costs the voted system 0.202 FnAcc and 0.190 ArgEM, and retraining one adapter per base on permuted candidates removes the dependence with no loss of ArgEM. One post-processing rule contributes 0.028 ArgEM and is the only component whose gain separates from zero, the other targets a convention the final data release removed and is inert, and the largest effect we can measure comes from the benchmark{'}s construction rather than from our modelling."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="alshehri-etal-2026-durham-jazan">
<titleInfo>
<title>Durham-Jazan at AISA-ArabicFC: A Field-Wise QLoRA Ensemble and the Sensitivity of Exact-Match Scoring</title>
</titleInfo>
<name type="personal">
<namePart type="given">Amal</namePart>
<namePart type="given">Saad</namePart>
<namePart type="family">Alshehri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nelly</namePart>
<namePart type="family">Bencomo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Housam</namePart>
<namePart type="family">Babiker</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amir</namePart>
<namePart type="family">Atapour-Abarghouei</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 Durham-Jazan submission to AISA-ArabicFC 2026, an Arabic function-calling task in which a system selects one tool from four candidates and fills its arguments, scored on tool choice (FnAcc) and on strict exact match of the whole argument dictionary (ArgEM). Across six QLoRA fine-tunes of ALLaM-7B and Qwen2.5-7B, trained on the provided data, negatives derived from it and 68 hand-written no-tool utterances, and run on one 16 GB GPU, we vote on each argument separately, then apply two deterministic rules estimated from training statistics. The system reaches ArgEM 0.824 on development under the final evaluator and 0.826 on the blind test, placing 9th of 17 in Track A, 8th of 17 in Track B and 11th of 21 in Track C. Our main findings concern the benchmark. The reference tool is listed first on every training and development row that calls one, in every data release. Shuffling the candidates at inference costs the voted system 0.202 FnAcc and 0.190 ArgEM, and retraining one adapter per base on permuted candidates removes the dependence with no loss of ArgEM. One post-processing rule contributes 0.028 ArgEM and is the only component whose gain separates from zero, the other targets a convention the final data release removed and is inert, and the largest effect we can measure comes from the benchmark’s construction rather than from our modelling.</abstract>
<identifier type="citekey">alshehri-etal-2026-durham-jazan</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.200/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1552</start>
<end>1563</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Durham-Jazan at AISA-ArabicFC: A Field-Wise QLoRA Ensemble and the Sensitivity of Exact-Match Scoring
%A Alshehri, Amal Saad
%A Bencomo, Nelly
%A Babiker, Housam
%A Atapour-Abarghouei, Amir
%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 alshehri-etal-2026-durham-jazan
%X We describe the Durham-Jazan submission to AISA-ArabicFC 2026, an Arabic function-calling task in which a system selects one tool from four candidates and fills its arguments, scored on tool choice (FnAcc) and on strict exact match of the whole argument dictionary (ArgEM). Across six QLoRA fine-tunes of ALLaM-7B and Qwen2.5-7B, trained on the provided data, negatives derived from it and 68 hand-written no-tool utterances, and run on one 16 GB GPU, we vote on each argument separately, then apply two deterministic rules estimated from training statistics. The system reaches ArgEM 0.824 on development under the final evaluator and 0.826 on the blind test, placing 9th of 17 in Track A, 8th of 17 in Track B and 11th of 21 in Track C. Our main findings concern the benchmark. The reference tool is listed first on every training and development row that calls one, in every data release. Shuffling the candidates at inference costs the voted system 0.202 FnAcc and 0.190 ArgEM, and retraining one adapter per base on permuted candidates removes the dependence with no loss of ArgEM. One post-processing rule contributes 0.028 ArgEM and is the only component whose gain separates from zero, the other targets a convention the final data release removed and is inert, and the largest effect we can measure comes from the benchmark’s construction rather than from our modelling.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.200/
%P 1552-1563
Markdown (Informal)
[Durham-Jazan at AISA-ArabicFC: A Field-Wise QLoRA Ensemble and the Sensitivity of Exact-Match Scoring](https://aclanthology.org/2026.arabicnlp-sharedtasks.200/) (Alshehri et al., ArabicNLP 2026)
ACL