AdaptLinker at KnowledgeGraphEval 2026: Cross-Domain Arabic Named Entity Recognition with Adversarial and Self-Training

Team AdaptLinker, represented by Quan Hoang Ho and Thìn Đặng Văn, ranked 6th in the AdaptNER shared task Subtask 1 (Cross-Domain Named Entity Recognition) at KnowledgeGraphEval 2026, reporting a Micro F1 of 73.29% and Macro F1 of 63.37% on the official blinded test set, per the paper published in the Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks (pages 1195–1199, Budapest, Hungary, Association for Computational Linguistics). The system's Precision was 75.48% and Recall 71.21%, beating the shared-task baseline of 62.2% Micro F1. The approach built on an AraBERT backbone combined 21 parallel linear classifiers for nested entities, a Domain Adversarial Neural Network with a Gradient Reversal Layer, confidence-gated pseudo-labeling on unlabeled target text, and Stochastic Weight Averaging to handle domain shift across 10 unseen target domains.

Abstract This paper presents the submission of team AdaptLinker to the AdaptNER shared task Subtask 1: Cross-Domain Named Entity Recognition at KnowledgeGraphEval 2026. To address nested entities and domain shift across 10 unseen target domains, we propose a multi-head token classification framework built on the AraBERT backbone. Our approach inte- grates: 1 21 parallel linear classifiers to pre- dict overlapping named entities; 2 a Domain Adversarial Neural Network DANN using a Gradient Reversal Layer GRL to align feature spaces; 3 confidence-gated pseudo-labeling on unlabeled target text; and 4 Stochastic Weight Averaging SWA . Our final system achieved a Precision of 75.48%, Recall of 71.21%, a Micro F1 of 73.29%, and a Macro F1 of 63.37% on the official blinded test set, outperforming the shared-task baseline 0.7329 vs. 0.622 Micro F1 and ranking 6th out of all participating teams - Anthology ID: - 2026.arabicnlp-sharedtasks.149 - Volume: - Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks https://aclanthology.org/volumes/2026.arabicnlp-sharedtasks/ - Month: - October - Year: - 2026 - Address: - Budapest, Hungary - Editors: - Muhammad Abdul-Mageed https://aclanthology.org/people/muhammad-abdul-mageed/ , Tamer Elsayed https://aclanthology.org/people/tamer-elsayed/ , Houda Bouamor https://aclanthology.org/people/houda-bouamor/unverified/ , Samhaa El-Beltagy https://aclanthology.org/people/samhaa-r-el-beltagy/ , Hend Al-Khalifa https://aclanthology.org/people/hend-al-khalifa/ , Ahmed Abdelali https://aclanthology.org/people/ahmed-abdelali/unverified/ , AbdelRahim Elmadany https://aclanthology.org/people/abdelrahim-elmadany/unverified/ , Abdellah El Mekki https://aclanthology.org/people/abdellah-el-mekki/ , Kareem Darwish https://aclanthology.org/people/kareem-darwish/ , Sakhar Alkhereyf https://aclanthology.org/people/sakhar-alkhereyf/ , Hamzah Luqman https://aclanthology.org/people/hamzah-luqman/ , Salam Khalifa https://aclanthology.org/people/salam-khalifa/ , Samar Magdy https://aclanthology.org/people/samar-magdy/unverified/ , Zaid Alyafeai https://aclanthology.org/people/zaid-alyafeai/ , Injy Hamed https://aclanthology.org/people/injy-hamed/unverified/ , Wajdi Zaghouani https://aclanthology.org/people/wajdi-zaghouani/ , Imed Zitouni https://aclanthology.org/people/imed-zitouni/unverified/ - Venue: - ArabicNLP https://aclanthology.org/venues/arabicnlp/ - SIG: - SIGARAB https://aclanthology.org/sigs/sigarab/ - Publisher: - Association for Computational Linguistics - Note: - Pages: - 1195–1199 - Language: - URL: - https://aclanthology.org/2026.arabicnlp-sharedtasks.149/ https://aclanthology.org/2026.arabicnlp-sharedtasks.149/ - DOI: - Cite ACL : - Quan Hoang Ho and Thìn Đặng Văn. 2026. AdaptLinker at KnowledgeGraphEval 2026: Cross-Domain Arabic Named Entity Recognition with Adversarial and Self-Training https://aclanthology.org/2026.arabicnlp-sharedtasks.149/ . In Proceedings of The Fourth Arabic Natural Language Processing Conference: Shared Tasks , pages 1195–1199, Budapest, Hungary. Association for Computational Linguistics. - Cite Informal : - AdaptLinker at KnowledgeGraphEval 2026: Cross-Domain Arabic Named Entity Recognition with Adversarial and Self-Training https://aclanthology.org/2026.arabicnlp-sharedtasks.149/ Ho & Văn, ArabicNLP 2026 - PDF: - https://aclanthology.org/2026.arabicnlp-sharedtasks.149.pdf https://aclanthology.org/2026.arabicnlp-sharedtasks.149.pdf