International Research Journal of Education and Technology

 The rapid advancement of artificial intelligence and natural language processing (NLP) technologies has created new opportunities to automate and enhance academic assessment processes. This paper presents Elio, an AI-powered question paper generator that leverages state-of-the-art NLP models to automatically produce high-quality, contextually relevant, and pedagogically sound examination questions from any academic or research topic. The system integrates transformer-based language models, domain-specific fine-tuning, and a structured generation pipeline to create diverse question types including conceptual, analytical, application-based, and critical-thinking questions. Experimental results demonstrate that Elio achieves a semantic relevance score of 91.4%, a human evaluator acceptance rate of 87.6%, and significantly reduces question paper preparation time by 78% compared to manual processes. This work bridges the gap between AI-assisted education and scalable assessment generation, with direct implications for universities, research institutions, and online learning platforms.

Keywords: Natural Language Processing, Question Generation, Transformer Models, AI in Education, Automated Assessment, BERT, GPT, Elio System