NLPIR 2026 Spcial Session 1
Topic: Agentic NLP for Low-Resource Language Education
Organizer(s): Dr. K. Pal Thamburaj, Nanyang Technological University, Singapore
Doctoral Researcher Mercedes Premalatha Ramesh, Nanyang Technological University, Singapore
Low-resource language education faces persistent challenges, including limited digital corpora, inadequate teaching tools, insufficient assessment resources, and a lack of language-specific artificial intelligence systems. Recent developments in agentic NLP, multi-agent systems, reinforcement learning, machine learning, and deep learning offer new possibilities for addressing these challenges.
This special session focuses on the design, development, and evaluation of intelligent systems that support teachers and learners of low-resource languages. It welcomes research on autonomous educational agents, multi-agent content validation, adaptive learning systems, personalised feedback, automated assessment, speech and pronunciation technologies, and human–AI collaborative teaching.
The session will bring together researchers from natural language processing, artificial intelligence, education, human–computer interaction, and learning sciences. Particular attention will be given to systems that preserve linguistic accuracy, cultural context, explainability, teacher oversight, and responsible AI use. It encourages technical studies, classroom evaluations, datasets, benchmarks, frameworks, and practical educational applications involving underrepresented languages.
Topics of Interest:
Agentic NLP for language education
Multi-agent AI systems for teaching and learning
Autonomous educational and tutoring agents
Reinforcement learning for adaptive teaching
Reinforcement learning for personalised learning pathways
Machine learning and deep learning for low-resource languages
Human–AI collaborative teaching and co-teaching systems
Generative AI for educational content development
Multi-agent validation of teaching and learning materials
Personalised feedback and learner modelling
Automated language assessment and exercise generation
Speech recognition and pronunciation assessment
Retrieval-augmented generation for educational resources
NLP for multilingual classrooms and code-switching
Morphology-aware NLP for low-resource languages
Explainable and trustworthy educational AI
Educational NLP datasets, benchmarks, and shared tasks
Evaluation of learning outcomes, teacher workload, and user trust
Ethical and culturally appropriate AI for language education