Abstract
Test-based learning is effective in fostering knowledge retention, but manually creating assessment questions remains time-consuming and limits personalized student practice. The advent of Large Language Models (LLMs) has introduced new possibilities for Automatic Question Generation (AQG). Motivated by this context, this survey provides a comprehensive overview of AQG using LLMs, focusing on educational applications. Following the PRISMA methodology, we reviewed 132 studies published between 2023 and 2025. Our contributions include a taxonomy of question types by response openness, an analysis of AQG efforts across knowledge fields, educational levels, evaluation strategies, and difficulty control. We also identify recurring challenges and research opportunities.
Funding Information
  • Postdoctoral Researcher Program (PPPD) at UNICAMP, São Paulo Research Foundation (2025/19399-8, 2026/01653-8, and 2025/10403-2)
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico (316067/2023-7, 301337/2025-0, and 400456/2025-7)
  • Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Finance Code 001)