@inproceedings{syulistyo-etal-2026-armyforner,
title = "{A}rmy{F}or{NER} at {K}nowledge{G}raph{E}val 2026: Multi-Stage Domain Adaptation and Self-Training Ensemble for {A}rabic Nested Named Entity Recognition",
author = "Syulistyo, Arie Rachmad and
Shoumi, Milyun Ni{'}ma and
Putra, Savero Athallah Hardiana and
Azizah, Aulia Resty",
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.160/",
pages = "1281--1286",
ISBN = "979-8-89176-497-2",
abstract = "This paper presents the ArmyForNER system for Subtask 1 of the KnowledgeGraphEval 2026 shared task on Arabic named entity recognition (NER). We address nested, cross-domain Arabic NER with a multi-stage pipeline built on ARBERTv2: (1) domain-adaptive pre-training (DAPT) via masked language modelling on unlabelled Konooz text, (2) fine-tuning with a multi-head BIO tagging architecture on the Wojood nested NER corpus, and (3) self-training on pseudo-labels produced by an intermediate DAPT ensemble over the Konooz domains. Soft-voting three self-trained checkpoints obtains 74.85 micro-F1 and 65.20 macro-F1 on the official test set, against a 63.00 / 44.00 base- line. Because both adaptation stages consume the blind Konooz test sentences, our system is listed among those that used Konooz during training, and its score is not comparable to systems that used Konooz for evaluation only. We state this setting explicitly and find that self-training gives consistent gains across seeds, whereas the benefit of DAPT is configuration- and seed-dependent."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="syulistyo-etal-2026-armyforner">
<titleInfo>
<title>ArmyForNER at KnowledgeGraphEval 2026: Multi-Stage Domain Adaptation and Self-Training Ensemble for Arabic Nested Named Entity Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Arie</namePart>
<namePart type="given">Rachmad</namePart>
<namePart type="family">Syulistyo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Milyun</namePart>
<namePart type="given">Ni’ma</namePart>
<namePart type="family">Shoumi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Savero</namePart>
<namePart type="given">Athallah</namePart>
<namePart type="given">Hardiana</namePart>
<namePart type="family">Putra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aulia</namePart>
<namePart type="given">Resty</namePart>
<namePart type="family">Azizah</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>This paper presents the ArmyForNER system for Subtask 1 of the KnowledgeGraphEval 2026 shared task on Arabic named entity recognition (NER). We address nested, cross-domain Arabic NER with a multi-stage pipeline built on ARBERTv2: (1) domain-adaptive pre-training (DAPT) via masked language modelling on unlabelled Konooz text, (2) fine-tuning with a multi-head BIO tagging architecture on the Wojood nested NER corpus, and (3) self-training on pseudo-labels produced by an intermediate DAPT ensemble over the Konooz domains. Soft-voting three self-trained checkpoints obtains 74.85 micro-F1 and 65.20 macro-F1 on the official test set, against a 63.00 / 44.00 base- line. Because both adaptation stages consume the blind Konooz test sentences, our system is listed among those that used Konooz during training, and its score is not comparable to systems that used Konooz for evaluation only. We state this setting explicitly and find that self-training gives consistent gains across seeds, whereas the benefit of DAPT is configuration- and seed-dependent.</abstract>
<identifier type="citekey">syulistyo-etal-2026-armyforner</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.160/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1281</start>
<end>1286</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ArmyForNER at KnowledgeGraphEval 2026: Multi-Stage Domain Adaptation and Self-Training Ensemble for Arabic Nested Named Entity Recognition
%A Syulistyo, Arie Rachmad
%A Shoumi, Milyun Ni’ma
%A Putra, Savero Athallah Hardiana
%A Azizah, Aulia Resty
%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 syulistyo-etal-2026-armyforner
%X This paper presents the ArmyForNER system for Subtask 1 of the KnowledgeGraphEval 2026 shared task on Arabic named entity recognition (NER). We address nested, cross-domain Arabic NER with a multi-stage pipeline built on ARBERTv2: (1) domain-adaptive pre-training (DAPT) via masked language modelling on unlabelled Konooz text, (2) fine-tuning with a multi-head BIO tagging architecture on the Wojood nested NER corpus, and (3) self-training on pseudo-labels produced by an intermediate DAPT ensemble over the Konooz domains. Soft-voting three self-trained checkpoints obtains 74.85 micro-F1 and 65.20 macro-F1 on the official test set, against a 63.00 / 44.00 base- line. Because both adaptation stages consume the blind Konooz test sentences, our system is listed among those that used Konooz during training, and its score is not comparable to systems that used Konooz for evaluation only. We state this setting explicitly and find that self-training gives consistent gains across seeds, whereas the benefit of DAPT is configuration- and seed-dependent.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.160/
%P 1281-1286
Markdown (Informal)
[ArmyForNER at KnowledgeGraphEval 2026: Multi-Stage Domain Adaptation and Self-Training Ensemble for Arabic Nested Named Entity Recognition](https://aclanthology.org/2026.arabicnlp-sharedtasks.160/) (Syulistyo et al., ArabicNLP 2026)
ACL