@inproceedings{lipinski-makarov-2026-imak-ai,
title = "i{M}ak {AI} Lab at {A}ra{S}eg Shared Task 2026: Greedy-Selected {A}rabic Encoder Ensembles with Structural Constraints for Sentence Segmentation",
author = "Lipinski, Artiom and
Makarov, Ilya",
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.114/",
pages = "927--932",
ISBN = "979-8-89176-497-2",
abstract = "We describe the iMak AI Lab submissions to the closed track of the AraSeg Shared Task 2026 on Arabic sentence segmentation, covering all four subtasks: Paragraph-Aware (PA), No-Paragraph (NP), No-Punctuation Paragraph-Aware (NOPNX-PA), and NoPunctuation No-Paragraph (NOPNX-NP). Our system treats segmentation as binary token classification with pretrained Arabic encoders finetuned only on the 128K-token AraSeg training split. The single-model recipe encodes documents in overlapping 512-token windows and trains for three times more epochs than the standard schedule, which is the largest source of improvement in our pipeline. Per-subtask ensembles are then built by greedy forward selection over a pool of seven encoder families, several seeds, and multi-variant joint-training variants. Decoding applies deterministic structural constraints and a decision threshold tuned on the development set; no open-test labels or leaderboard feedback enter any modeling decision. The systems reach blind-test F1 scores of 94.1 (PA), 91.0 (NP), 89.5 (NOPNX-PA), and 85.8 (NOPNX-NP), above the official baseline on all four subtasks. We report ablations for each design decision, including several that did not help."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lipinski-makarov-2026-imak-ai">
<titleInfo>
<title>iMak AI Lab at AraSeg Shared Task 2026: Greedy-Selected Arabic Encoder Ensembles with Structural Constraints for Sentence Segmentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Artiom</namePart>
<namePart type="family">Lipinski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ilya</namePart>
<namePart type="family">Makarov</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 iMak AI Lab submissions to the closed track of the AraSeg Shared Task 2026 on Arabic sentence segmentation, covering all four subtasks: Paragraph-Aware (PA), No-Paragraph (NP), No-Punctuation Paragraph-Aware (NOPNX-PA), and NoPunctuation No-Paragraph (NOPNX-NP). Our system treats segmentation as binary token classification with pretrained Arabic encoders finetuned only on the 128K-token AraSeg training split. The single-model recipe encodes documents in overlapping 512-token windows and trains for three times more epochs than the standard schedule, which is the largest source of improvement in our pipeline. Per-subtask ensembles are then built by greedy forward selection over a pool of seven encoder families, several seeds, and multi-variant joint-training variants. Decoding applies deterministic structural constraints and a decision threshold tuned on the development set; no open-test labels or leaderboard feedback enter any modeling decision. The systems reach blind-test F1 scores of 94.1 (PA), 91.0 (NP), 89.5 (NOPNX-PA), and 85.8 (NOPNX-NP), above the official baseline on all four subtasks. We report ablations for each design decision, including several that did not help.</abstract>
<identifier type="citekey">lipinski-makarov-2026-imak-ai</identifier>
<location>
<url>https://aclanthology.org/2026.arabicnlp-sharedtasks.114/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>927</start>
<end>932</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T iMak AI Lab at AraSeg Shared Task 2026: Greedy-Selected Arabic Encoder Ensembles with Structural Constraints for Sentence Segmentation
%A Lipinski, Artiom
%A Makarov, Ilya
%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 lipinski-makarov-2026-imak-ai
%X We describe the iMak AI Lab submissions to the closed track of the AraSeg Shared Task 2026 on Arabic sentence segmentation, covering all four subtasks: Paragraph-Aware (PA), No-Paragraph (NP), No-Punctuation Paragraph-Aware (NOPNX-PA), and NoPunctuation No-Paragraph (NOPNX-NP). Our system treats segmentation as binary token classification with pretrained Arabic encoders finetuned only on the 128K-token AraSeg training split. The single-model recipe encodes documents in overlapping 512-token windows and trains for three times more epochs than the standard schedule, which is the largest source of improvement in our pipeline. Per-subtask ensembles are then built by greedy forward selection over a pool of seven encoder families, several seeds, and multi-variant joint-training variants. Decoding applies deterministic structural constraints and a decision threshold tuned on the development set; no open-test labels or leaderboard feedback enter any modeling decision. The systems reach blind-test F1 scores of 94.1 (PA), 91.0 (NP), 89.5 (NOPNX-PA), and 85.8 (NOPNX-NP), above the official baseline on all four subtasks. We report ablations for each design decision, including several that did not help.
%U https://aclanthology.org/2026.arabicnlp-sharedtasks.114/
%P 927-932
Markdown (Informal)
[iMak AI Lab at AraSeg Shared Task 2026: Greedy-Selected Arabic Encoder Ensembles with Structural Constraints for Sentence Segmentation](https://aclanthology.org/2026.arabicnlp-sharedtasks.114/) (Lipinski & Makarov, ArabicNLP 2026)
ACL