@inproceedings{alamr-etal-2026-thakaa-alexandriax,
title = "Thakaa at {A}lexandria{X}-2026: Who Gets the Final Say? Assigning Output Authority in Dialectal {A}rabic Machine Translation and Error Detection",
author = "Alamr, Meshal and
Altuwaijri, Maryam and
Alqaeri, Hassan and
Aldahlawi, Abdullah",
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.109/",
pages = "872--893",
ISBN = "979-8-89176-497-2",
abstract = "We report first-place systems for all four evaluated AlexandriaX-2026 settings: both English-to-dialect translation tracks, cross-dialect translation, and translation-error detection. In unconstrained translation, our minimum Bayes risk (MBR) selection restricts output authority: four local systems can supply the translation, while seven additional systems contribute only reference evidence. Candidate-constrained MBR beats the strongest complete candidate in all 13 dialects and gains 1.2554 macro spBLEU overall, reaching 33.4891. It also surpasses a retrospective oracle selecting the best whole candidate system per dialect. In the constrained track, country-routed LoRA adapters over a shared NileChat-3B-Base backbone incorporate the released Libyan and Sudanese supervision absent from the training and development splits, raising macro spBLEU from 29.5969 to 30.4233. Post-evaluation, a single adapter trained on all released Alexandria data reaches 30.4733 macro spBLEU; averaging model weights across three training seeds raises this to 30.9726 while retaining one model for inference. For cross-dialect translation, nonparallel target-dialect retrieval guides deliberation, raising Gemma-4-31B-IT without fine-tuning from 24.6898 to 28.4011 spBLEU. For translation-error detection, GPT-5.5 proposes semantic error regions, while MARBERTv2 controls their character boundaries and corroborates additional spans, raising Overall from 0.4691 to 0.4982."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="alamr-etal-2026-thakaa-alexandriax">
<titleInfo>
<title>Thakaa at AlexandriaX-2026: Who Gets the Final Say? Assigning Output Authority in Dialectal Arabic Machine Translation and Error Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Meshal</namePart>
<namePart type="family">Alamr</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maryam</namePart>
<namePart type="family">Altuwaijri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hassan</namePart>
<namePart type="family">Alqaeri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdullah</namePart>
<namePart type="family">Aldahlawi</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 report first-place systems for all four evaluated AlexandriaX-2026 settings: both English-to-dialect translation tracks, cross-dialect translation, and translation-error detection. In unconstrained translation, our minimum Bayes risk (MBR) selection restricts output authority: four local systems can supply the translation, while seven additional systems contribute only reference evidence. Candidate-constrained MBR beats the strongest complete candidate in all 13 dialects and gains 1.2554 macro spBLEU overall, reaching 33.4891. It also surpasses a retrospective oracle selecting the best whole candidate system per dialect. In the constrained track, country-routed LoRA adapters over a shared NileChat-3B-Base backbone incorporate the released Libyan and Sudanese supervision absent from the training and development splits, raising macro spBLEU from 29.5969 to 30.4233. Post-evaluation, a single adapter trained on all released Alexandria data reaches 30.4733 macro spBLEU; averaging model weights across three training seeds raises this to 30.9726 while retaining one model for inference. For cross-dialect translation, nonparallel target-dialect retrieval guides deliberation, raising Gemma-4-31B-IT without fine-tuning from 24.6898 to 28.4011 spBLEU. For translation-error detection, GPT-5.5 proposes semantic error regions, while MARBERTv2 controls their character boundaries and corroborates additional spans, raising Overall from 0.4691 to 0.4982.</abstract>
<identifier type="citekey">alamr-etal-2026-thakaa-alexandriax</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.109/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>872</start>
<end>893</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Thakaa at AlexandriaX-2026: Who Gets the Final Say? Assigning Output Authority in Dialectal Arabic Machine Translation and Error Detection
%A Alamr, Meshal
%A Altuwaijri, Maryam
%A Alqaeri, Hassan
%A Aldahlawi, Abdullah
%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 alamr-etal-2026-thakaa-alexandriax
%X We report first-place systems for all four evaluated AlexandriaX-2026 settings: both English-to-dialect translation tracks, cross-dialect translation, and translation-error detection. In unconstrained translation, our minimum Bayes risk (MBR) selection restricts output authority: four local systems can supply the translation, while seven additional systems contribute only reference evidence. Candidate-constrained MBR beats the strongest complete candidate in all 13 dialects and gains 1.2554 macro spBLEU overall, reaching 33.4891. It also surpasses a retrospective oracle selecting the best whole candidate system per dialect. In the constrained track, country-routed LoRA adapters over a shared NileChat-3B-Base backbone incorporate the released Libyan and Sudanese supervision absent from the training and development splits, raising macro spBLEU from 29.5969 to 30.4233. Post-evaluation, a single adapter trained on all released Alexandria data reaches 30.4733 macro spBLEU; averaging model weights across three training seeds raises this to 30.9726 while retaining one model for inference. For cross-dialect translation, nonparallel target-dialect retrieval guides deliberation, raising Gemma-4-31B-IT without fine-tuning from 24.6898 to 28.4011 spBLEU. For translation-error detection, GPT-5.5 proposes semantic error regions, while MARBERTv2 controls their character boundaries and corroborates additional spans, raising Overall from 0.4691 to 0.4982.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.109/
%P 872-893
Markdown (Informal)
[Thakaa at AlexandriaX-2026: Who Gets the Final Say? Assigning Output Authority in Dialectal Arabic Machine Translation and Error Detection](https://aclanthology.org/2026.arabicnlp-sharedtasks.109/) (Alamr et al., ArabicNLP 2026)
ACL