Embedded Files
Program
Saturday November 8
Saturday November 8
(8:30 - 8:45) Welcome Speech & State of SIGARAB
(8:30 - 8:45) Welcome Speech & State of SIGARAB
(8:45 - 09:30) Invited Talk by Houda Bouamor: “Beyond Resources: Building an Arabic NLP Ecosystem Rooted in Representation, Collaboration, and Responsibility”
(8:45 - 09:30) Invited Talk by Houda Bouamor: “Beyond Resources: Building an Arabic NLP Ecosystem Rooted in Representation, Collaboration, and Responsibility”
(9:30 - 10:30) Session 1: LLM Benchmarking & Development (1)
(9:30 - 10:30) Session 1: LLM Benchmarking & Development (1)
(09:30 - 09:45) - AraHalluEval: A Fine-grained Hallucination Evaluation Framework for Arabic LLMs
Aisha Alansari and Hamzah Luqman(09:45 - 10:00) - BALSAM: A Platform for Benchmarking Arabic Large Language Models
Rawan Nasser Almatham, Kareem Mohamed Darwish, Raghad AlRasheed, Waad Thuwaini Alshammari, Muneera Alhoshan, Amal Almazrua, Asma Al Wazrah, Mais Alheraki, Firoj Alam, Preslav Nakov, Norah A. Alzahrani, Eman Albilali, Nizar Habash, Abdelrahman Mustafa ElSheikh, Muhammad Elmallah, Hamdy Mubarak, Zaid Alyafeai, Mohamed Anwar, Haonan Li, Ahmed Abdelali, Nora Altwairesh, Maram Hasanain, Abdulmohsen AlThubaity, Shady Shehata, Bashar Alhafni, Injy Hamed, Go Inoue, Khalid N. Elmadani, Ossama Obeid, Fatima Haouari, Tamer Elsayed, Emad A. Alghamdi, Khalid Almubarak, Saied Alshahrani, Ola Aljareh, Safa Alajlan, Areej Alshaqarawi, Maryam Alshihri, Sultana Alghurabi, Atikah Alzeghayer, Afrah Altamimi, Abdullah Alfaifi and Abdulrahman M Alosaimy(10:00 - 10:15) - 3LM: Bridging Arabic, STEM, and Code through Benchmarking
Basma El Amel Boussaha, Leen Al Qadi, Mugariya Farooq, Shaikha Alsuwaidi, Giulia Campesan, Ahmed Alzubaidi, Mohammed Alyafeai and Hakim Hacid(10:15 - 10:30) - Nile-Chat: Egyptian Language Models for Arabic and Latin Scripts
Guokan Shang, Hadi Abdine, Ahmad Chamma, Amr Mohamed, Mohamed Anwar, Abdelaziz Bounhar, Omar El Herraoui, Preslav Nakov, Michalis Vazirgiannis and Eric P. Xing
(10:30 - 11:00) Coffee Break
(10:30 - 11:00) Coffee Break
(11:00 - 11:30) Session 2: LLM Benchmarking & Development (2)
(11:00 - 11:30) Session 2: LLM Benchmarking & Development (2)
(11:00 - 11:15) - Mind the Gap: A Review of Arabic Post-Training Datasets and Their Limitations
Mohammed Alkhowaiter, Saied Alshahrani, Norah F Alshahrani, Reem I. Masoud, Alaa Alzahrani, Deema Alnuhait, Emad A. Alghamdi and Khalid Almubarak(11:15 - 11:30) - Adapting Falcon3-7B Language Model for Arabic: Methods, Challenges, and Outcomes
Basma El Amel Boussaha, Mohammed Alyafeai, Ahmed Alzubaidi, Leen Al Qadi, Shaikha Alsuwaidi and Hakim Hacid
(11:30 - 12:30) Session 3: Arabic Resources, Modeling, and Representation
(11:30 - 12:30) Session 3: Arabic Resources, Modeling, and Representation
(11:30 - 11:45) - Capturing Intra-Dialectal Variation in Qatari Arabic: A Corpus of Cultural and Gender Dimensions
Houda Bouamor, Sara Al-Emadi, Zeinab Ibrahim, Hany Fazzaa and Aisha Al-Sultan(11:45 - 12:00) - Lemmatizing Dialectal Arabic with Sequence-to-Sequence Models
Mostafa Saeed and Nizar Habash(12:00 - 12:15) - Semitic Root Encoding: Tokenization Based on the Templatic Morphology of Semitic Languages in NMT
Brendan T. Hatch and Stephen D. Richardson(12:15 - 12:30) - Learning Word Embeddings from Glosses: A Multi-Loss Framework for Arabic Reverse Dictionary Tasks
Engy Ibrahim, Farhah Adel, Marwan Torki and Nagwa El-Makky
(12:30 - 14:00) - Lunch Break + SIGARAB Business Meeting
(12:30 - 14:00) - Lunch Break + SIGARAB Business Meeting
(14:00 - 14:30) Session 4: Multi-Modality
(14:00 - 14:30) Session 4: Multi-Modality
(14:00 - 14:15) - AMCrawl: An Arabic Web-Scale Dataset of Interleaved Image-Text Documents and Image-Text Pairs
Shahad Aboukozzana, Muhammad Kamran J Khan and Ahmed Ali(14:15 - 14:30) - TuniFra: A Tunisian Arabic Speech Corpus with Orthographic Transcriptions and French Translations
Alex Choux, Marko Avila, Josep Crego, Fethi Bougares and Antoine Laurent
(14:30 - 15:30) Shared Task Overviews (1)
(14:30 - 15:30) Shared Task Overviews (1)
(14:30 - 14:40) - Shared Task 1: ImageEval Arabic Image Captioning
(14:40 - 14:50) - Shared Task 2: Iqra'Eval: Qur'anic Pronunciation Assessment
(14:50 - 15:00) - Shared Task 3: NADI 2025: Multidialectal Arabic Speech Processing
(15:00 - 15:10) - Shared Task 4: MAHED 2025: Multimodal Detection of Hope and Hate Emotions in Arabic Content
(15:10 - 15:20) - Shared Task 5: AraGenEval: Arabic Authorship Style Transfer and AI Generated Text Detection
(15:20 - 15:30) - Shared Task 6: TAQEEM 2025: The First Task for Arabic Quality Evaluation of Essays in Multi-dimensions
(15:30 - 16:00) - Coffee Break
(15:30 - 16:00) - Coffee Break
(16:00 - 17:00) Shared Task Overviews (2)
(16:00 - 17:00) Shared Task Overviews (2)
(16:00 - 16:10) - Shared Task 7: BAREC 2025: Arabic Readability Assessment Shared Task
(16:10 - 16:20) - Shared Task 8: AraHealthQA 2025: Comprehensive Arabic Health Question Answering
(16:20 - 16:30) - Shared Task 9: IslamicEval: Capturing LLMs Hallucination in Islamic Content
(16:30 - 16:40) - Shared Task 10: PalmX 2025: The First Shared Task on Benchmarking LLMs on Arabic Culture
(16:40 - 16:50) - Shared Task 11: QIAS 2025: Q&A in Islamic Studies Assessment
(17:00 - 18:00) Poster Session
(17:00 - 18:00) Poster Session
Saudi-Alignment Benchmark: Assessing LLMs Alignment with Cultural Norms and Domain Knowledge in the Saudi Context
Manal Alhassoun, Imaan Mohammed Alkhanen, Nouf Alshalawi, Ibtehal Baazeem and Waleed AlsanieZero-Shot and Fine-Tuned Evaluation of Generative LLMs for Arabic Word Sense Disambiguation
Yossra Noureldien, Abdelrazig Mohamed and Farah AttallahModeling North African Dialects from Standard Languages
Yassine Toughrai, Kamel Sma¨ ıli and David LangloisArabicWeb-Edu: Educational Quality Data for Arabic LLM Training
Majd Hawasly, Tasnim Mohiuddin, Hamdy Mubarak and Sabri BoughorbelATHAR: A High-Quality and Diverse Dataset for Classical Arabic to English Translation
Mohammed Sabry Mohammed and Mohammed KhalilTahḏīb: A Rhythm-Aware Phrase Insertion for Classical Arabic Poetry Composition
Mohamad Elzohbi and Richard ZhaoTransfer or Translate? Argument Mining in Arabic with No Native Annotations
Sara Nabhani and Khalid Al KhatibInExOntology - Ontology-Driven LLM Prompting for Unified Information Extraction Tasks
Alaa Aljabari, Nagham Hamad, Mohammed Khalilia and Mustafa JarrarBridging Dialectal Gaps in Arabic Medical LLMs through Model Merging
Ahmed Ibrahim, Abdullah Hosseini, Hoda Helmy, Wafa Lakhdhar and Ahmed SeragTool Calling for Arabic LLMs: Data Strategies and Instruction Tuning
Asım Ersoy, Enes Altinisik, Kareem Mohamed Darwish and Husrev Taha SencarAn Exploration of Knowledge Editing for Arabic
Basel Mousi, Nadir Durrani and Fahim DalviAutoArabic: A Three-Stage Framework for Localizing Video-Text Retrieval Benchmarks
Mohamed Eltahir, Osamah Sarraj, Abdulrahman M. Alfrihidi, Taha Alshatiri, Mohammed Khurd, Mohammed Bremoo and Tanveer HussainOpen-domain Arabic Conversational Question Answering with Question Rewriting
Mariam E. Hassib, Nagwa El-Makky and Marwan TorkiDialG2P: Dialectal Grapheme-to-Phoneme. Arabic as a Case Study
Majd Hawasly, Hamdy Mubarak, Ahmed Abdelali and Ahmed AliThe Cross-Lingual Cost: Retrieval Biases in RAG over Arabic-English Corpora
Chen Amiraz, Yaroslav Fyodorov, Elad Haramaty, Zohar Karnin and Liane Lewin-Eytan
(17:00 - 18:00) - Shared Task Papers (Posters)
(17:00 - 18:00) - Shared Task Papers (Posters)
Sunday November 9
Sunday November 9
(8:30 - 8:45) Welcome Note
(8:30 - 8:45) Welcome Note
(8:45 - 09:30) Invited Talk by Areeb Alowisheq: “From Benchmarks to the Real-World Impact: Arabic LLMs in Production”
(8:45 - 09:30) Invited Talk by Areeb Alowisheq: “From Benchmarks to the Real-World Impact: Arabic LLMs in Production”
(9:30 - 10:30) Round Table (1)
(9:30 - 10:30) Round Table (1)
(10:30 - 11:00) Coffee Break
(10:30 - 11:00) Coffee Break
(11:00 - 12:30) Session 1: Education and Speech (2)
(11:00 - 12:30) Session 1: Education and Speech (2)
(11:00 - 11:15) - ArabJobs: A Multinational Corpus of Arabic Job Ads
Mo El-Haj(11:15 - 11:30) - Feature Engineering is not Dead: A Step Towards State of the Art for Arabic Automated Essay Scoring
Marwan Sayed, Sohaila Eltanbouly, May Bashendy and Tamer Elsayed(11:30 - 11:45) - Evaluating Prompt Relevance in Arabic Automatic Essay Scoring: Insights from Synthetic and Real-World Data
Chatrine Qwaider, Kirill Chirkunov, Bashar Alhafni, Nizar Habash and Ted Briscoe(11:45 - 12:00) - TEDxTN: : Three-way Speech Translation Corpus for Code-Switched Tunisian Arabic - English
Fethi Bougares, Salima Mdhaffar, Haroun Elleuch and Yannick Estève(12:00 - 12:15) - Octopus: Towards Building the Arabic Speech LLM Suite
Sara Althubaiti, Vasista Sai Lodagala, Tjad Clark, Yousseif Ahmed Elshahawy, Daniel Izham, Abdullah Alrajeh, Aljawahrah Bin Tamran and Ahmed Ali(12:15 - 12:30) - ArabEmoNet: A Lightweight Hybrid 2D CNN-BiLSTM Model with Attention for Robust Arabic Speech Emotion Recognition
Ali Abouzeid, Bilal Elbouardi, Mohamed Maged and Shady Shehata
(12:30 - 14:00) - Lunch Break
(12:30 - 14:00) - Lunch Break
(14:00 - 14:30) Session 2: Legal & Agents
(14:00 - 14:30) Session 2: Legal & Agents
(14:00 - 14:15) - ALARB: An Arabic Legal Argument Reasoning Benchmark
Harethah Abu Shairah, Somayah S. Alharbi, Abdulaziz A. AlHussein, Sameer Alsabea, Omar Shaqaqi, Hebah A. Alshamlan, Omar Knio and George Turkiyyah(14:15 - 14:30) - A-SEA3L-DU: An Fully Automated Self-Evolving, Adversarial Agentic Framework for Arabic Long-Context Document Understanding
Kesen Wang, Daulet Toibazar and Pedro J Moreno Mengibar
(14:30 - 15:30) Arab Culture & Retrieval
(14:30 - 15:30) Arab Culture & Retrieval
(14:30 - 14:45) - Assessing Large Language Models on Islamic Legal Reasoning: Evidence from Inheritance Law Evaluation
Abdessalam Bouchekif, Samer Rashwani, Heba Sbahi, Shahd Gaben, Mutaz Al Khatib and Mohammed Ghaly(14:45 - 15:00) - Toward Culturally-Aware Arabic Debate Platforms with NLP Support
Khalid Al Khatib and Mohammad Khader(15:00 - 15:15) - Can LLMs Directly Retrieve Passages for Answering Questions from Qur'an?
Sohaila Eltanbouly, Salam Albatarni, Shaimaa Hassanein and Tamer Elsayed(15:15 - 15:30) - Shawarma Chats: A Benchmark Exact Dialogue & Evaluation Platter in Egyptian, Maghrebi & Modern Standard Arabic—A Triple-Dialect Feast for Hungry Language Models
Kamyar Zeinalipour, Mohamed Zaky Saad, Oumaima Attafi, Marco Maggini and Marco Gori
(15:30 - 16:00) - Coffee Break
(15:30 - 16:00) - Coffee Break
(16:00 - 17:00) Round Table
(16:00 - 17:00) Round Table
Invited Talk (Nov 8): Beyond Resources: Building an Arabic NLP Ecosystem Rooted in Representation, Collaboration, and Responsibility
Invited Talk (Nov 8): Beyond Resources: Building an Arabic NLP Ecosystem Rooted in Representation, Collaboration, and Responsibility
Speaker: Dr. Houda Bouamor, CMU Qatar
Bio: Houda is an Associate Teaching Professor and Associate Area Head of the Information Systems Department at CMU-Q. She is also affiliated with the CAMeL Lab (Computational Approaches to Modeling Language Lab) at NYU Abu Dhabi, where she collaborates with Nizar Habash on several projects. Her work bridges applied machine learning, natural language processing, and generative AI, with a special focus on Arabic dialects, low-resource languages, and educational technologies.
Houda's area of research is Artificial Intelligence, specifically Natural Language Processing and Computational Linguistics. She primarily focuses on Arabic and Arabic dialect processing across orthography, morphology, syntax, semantics, lexicons, and corpora. She also works on machine translation, dialogue systems, and multilingual AI. A strong theme in my research is using AI for social good, ensuring that technology contributes to equity, inclusion, and meaningful societal impact.
Houda earned her PhD in Computer Science (Computational Linguistics) from Paris-Sud University, France, where she worked on paraphrase alignment under the supervision of Anne Vilnat and Aurélien Max. Before that, she completed her M.Sc. in Computer Science at the Paris-Est Marne-La-Vallée University and her B.Sc. in Computer Science at the University of Manouba, Tunisia.
Invited Talk (Nov 9): From Benchmarks to the Real-World Impact: Arabic LLMs in Production
Invited Talk (Nov 9): From Benchmarks to the Real-World Impact: Arabic LLMs in Production
Speaker: Dr. Areeb Alowisheq, HUMAIN
Bio: Areeb Alowisheq focuses on developing and managing research projects to build competing Arabic Language technologies. As Vice President of AI Research at HUMAIN and Head of HUMAIN Chat, she leads efforts to develop human-aligned generative and agentic technologies. Formally Assistant CEO for Research and Development at the National Center for AI at SDAIA, she leads the training of ALLAM and previously SauTech programs, Saudi Arabia’s flagship LLM and speech initiatives. Previously an Assistant Professor of Computer Science at Imam University, Areeb’s work bridges research, productization, and governance to advance a sustainable Arabic AI ecosystem.
Google Sites
Report abuse