Dr Mo El-Haj
Dr Mo El-Haj leads NLP @ VinUniversity and works on multilingual and low-resource NLP, Arabic language technology, text summarisation, information extraction, financial NLP, language resources and large language models.
Researching language technologies for multilingual, low-resource and real-world settings.
We develop Natural Language Processing methods, datasets, language resources and AI systems spanning multilingual NLP, large language models, text summarisation, information extraction and applications in areas including finance, healthcare, education and public policy.
NLP @ VinUniversity is the Natural Language Processing research group at VinUniversity. Our work combines core NLP research with multilingual and low-resource language technology, open language resources and applied AI.
Our research asks how language technologies can become more capable, reliable and useful beyond the small number of languages and domains that dominate mainstream AI development. This includes building datasets and benchmarks, developing new NLP methods, adapting large language models, and evaluating systems in linguistically and culturally diverse settings.
The group works across computational linguistics, machine learning and generative AI. We are particularly interested in multilingual NLP, under-resourced languages, language resources, text summarisation, information extraction, machine translation, retrieval and reasoning, as well as responsible evaluation of modern language models.
We collaborate with academic institutions, research communities, government and industry partners internationally, while developing research programmes that connect VinUniversity to the wider NLP community.
Dr Mo El-Haj leads NLP @ VinUniversity and works on multilingual and low-resource NLP, Arabic language technology, text summarisation, information extraction, financial NLP, language resources and large language models.
A flexible directory for faculty, researchers, research assistants, PhD students, visiting researchers and collaborators associated with the group.
Monash University, Australia
Lancaster University, UK
Cardiff University, UK
Autonomous University of Madrid, Spain
VinUniversity, Vietnam
VinUniversity, Vietnam
University of Southern Denmark, Denmark
VinUniversity, Vietnam
VinUniversity, Vietnam
HBKU, Qatar
VinUniversity, Vietnam
KFUP, Saudi Arabia
VinUniversity, Vietnam
VinUniversity, Vietnam
VinUniversity, Vietnam
VinUniversity, Vietnam
The member cards are intentionally prepared as editable placeholders. Replace the name, affiliation and image path for each person as the group directory is updated.
Our research combines foundational NLP with multilingual resources, generative AI and applied language technology.
Language modelling, transfer, evaluation and resource development for languages and varieties that remain under-represented in NLP.
Instruction tuning, adaptation, retrieval-augmented generation, multi-agent systems, model evaluation and reasoning.
Corpora, benchmarks, datasets, lexicons, annotation frameworks and open-source NLP tools.
Document understanding, automatic summarisation, information extraction and structured analysis of complex text.
Sentiment, classification, question answering, retrieval, human evaluation and robust multilingual benchmarking.
Language technology for public policy, finance, healthcare, education, social sciences and cultural heritage.
Current and recent projects illustrate the group's emphasis on multilingual AI, open resources, low-resource language technology and real-world NLP.
PolicyVerse develops multilingual retrieval, policy world models and evidence-grounded multi-agent reasoning for analysing complex policy information across languages.
Explore PolicyVersePolyDrift investigates how multilingual instruction-tuned language models change their language behaviour after adaptation, including when models drift away from the language requested by users.
Explore PolyDriftA benchmarked Vietnamese-English toolkit for segmentation, sentiment, summarisation and free-text corpus exploration.
View projectResearch and resource development for semantic analysis and tagging of Vietnamese text.
View projectSupporting bilingual free-text survey and questionnaire analysis through corpus-based and NLP methods.
Visit FreeTxtThe group works with research organisations and universities internationally across multilingual NLP, language resources and applied AI.
For research collaboration, projects, doctoral study and enquiries about NLP @ VinUniversity.
College of Engineering & Computer Science
VinUniversity, Hanoi, Vietnam
Email: elhaj.m@vinuni.edu.vn
Website: vinnlp.com