@inproceedings{saadaoui-etal-2026-quality,
title = "Quality-Aware {LLM} Data Augmentation for {A}rabic Dialectal {NER}",
author = "Saadaoui, Kahina Houda and
Abdelaziz, Chaima and
Belbachir, Faiza and
Said Lhadj, Lynda",
editor = "Wang, Yimu and
Choi, Yee Man and
Wu, Di and
Ouyang, Siqi and
Azadani, Mozhgan Nasr and
Fung, Yi R. (May)",
booktitle = "Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently ({G}round{LM} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.groundlm-1.50/",
pages = "666--684",
ISBN = "979-8-89176-478-1",
abstract = "Named Entity Recognition for dialectal Arabic remains severely constrained by data scarcity, and LLM-based augmentation is the natural remedy. We study LLM-based augmentation for the Lebanese and Palestinian sub-corpora of Wojood 2024, using LLaMA-3.1-8B-Instruct as a generation backbone, and compare three strategies (entity-level, context-level, combined) across two k-shot sampling paradigms. The proposed multi-dimensional pipeline yields a per-sentence composite score Q(s) that exposes a sharp dialectal-authenticity gap between entity- and context-level rewriting (DA 0.92 vs. 0.51). Despite lower register fidelity, context augmentation with curriculum training delivers the most consistent gains (Curras F1 +5.70 pp); entity augmentation lifts Lebanese F1 from a zero baseline to 50.00 on Lebanese-only training, a figure the 85-sentence test set makes coarse. Ablation, weight-sensitivity, threshold sweep, a random-selection control, and a second LLM backbone show that the dialectal-authenticity dimension carries most of Q(s){'}s protective effect on Lebanese and that generator choice is itself a first-class register-preservation lever."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="saadaoui-etal-2026-quality">
<titleInfo>
<title>Quality-Aware LLM Data Augmentation for Arabic Dialectal NER</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kahina</namePart>
<namePart type="given">Houda</namePart>
<namePart type="family">Saadaoui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chaima</namePart>
<namePart type="family">Abdelaziz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Faiza</namePart>
<namePart type="family">Belbachir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lynda</namePart>
<namePart type="family">Said Lhadj</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 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yimu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yee</namePart>
<namePart type="given">Man</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siqi</namePart>
<namePart type="family">Ouyang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mozhgan</namePart>
<namePart type="given">Nasr</namePart>
<namePart type="family">Azadani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="given">R</namePart>
<namePart type="given">(May)</namePart>
<namePart type="family">Fung</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">979-8-89176-478-1</identifier>
</relatedItem>
<abstract>Named Entity Recognition for dialectal Arabic remains severely constrained by data scarcity, and LLM-based augmentation is the natural remedy. We study LLM-based augmentation for the Lebanese and Palestinian sub-corpora of Wojood 2024, using LLaMA-3.1-8B-Instruct as a generation backbone, and compare three strategies (entity-level, context-level, combined) across two k-shot sampling paradigms. The proposed multi-dimensional pipeline yields a per-sentence composite score Q(s) that exposes a sharp dialectal-authenticity gap between entity- and context-level rewriting (DA 0.92 vs. 0.51). Despite lower register fidelity, context augmentation with curriculum training delivers the most consistent gains (Curras F1 +5.70 pp); entity augmentation lifts Lebanese F1 from a zero baseline to 50.00 on Lebanese-only training, a figure the 85-sentence test set makes coarse. Ablation, weight-sensitivity, threshold sweep, a random-selection control, and a second LLM backbone show that the dialectal-authenticity dimension carries most of Q(s)’s protective effect on Lebanese and that generator choice is itself a first-class register-preservation lever.</abstract>
<identifier type="citekey">saadaoui-etal-2026-quality</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.50/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>666</start>
<end>684</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Quality-Aware LLM Data Augmentation for Arabic Dialectal NER
%A Saadaoui, Kahina Houda
%A Abdelaziz, Chaima
%A Belbachir, Faiza
%A Said Lhadj, Lynda
%Y Wang, Yimu
%Y Choi, Yee Man
%Y Wu, Di
%Y Ouyang, Siqi
%Y Azadani, Mozhgan Nasr
%Y Fung, Yi R. (May)
%S Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-478-1
%F saadaoui-etal-2026-quality
%X Named Entity Recognition for dialectal Arabic remains severely constrained by data scarcity, and LLM-based augmentation is the natural remedy. We study LLM-based augmentation for the Lebanese and Palestinian sub-corpora of Wojood 2024, using LLaMA-3.1-8B-Instruct as a generation backbone, and compare three strategies (entity-level, context-level, combined) across two k-shot sampling paradigms. The proposed multi-dimensional pipeline yields a per-sentence composite score Q(s) that exposes a sharp dialectal-authenticity gap between entity- and context-level rewriting (DA 0.92 vs. 0.51). Despite lower register fidelity, context augmentation with curriculum training delivers the most consistent gains (Curras F1 +5.70 pp); entity augmentation lifts Lebanese F1 from a zero baseline to 50.00 on Lebanese-only training, a figure the 85-sentence test set makes coarse. Ablation, weight-sensitivity, threshold sweep, a random-selection control, and a second LLM backbone show that the dialectal-authenticity dimension carries most of Q(s)’s protective effect on Lebanese and that generator choice is itself a first-class register-preservation lever.
%U https://aclanthology.org/2026.groundlm-1.50/
%P 666-684
Markdown (Informal)
[Quality-Aware LLM Data Augmentation for Arabic Dialectal NER](https://aclanthology.org/2026.groundlm-1.50/) (Saadaoui et al., GroundLM 2026)
ACL
- Kahina Houda Saadaoui, Chaima Abdelaziz, Faiza Belbachir, and Lynda Said Lhadj. 2026. Quality-Aware LLM Data Augmentation for Arabic Dialectal NER. In Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026), pages 666–684, Budapest, Hungary. Association for Computational Linguistics.