From Memorization to Creation: Evaluating the Cognitive Depth of LLM-Generated Educational Questions
A new research paper evaluates six large language models (LLMs) on their ability to generate educational questions that go beyond simple memorization, using Bloom's Taxonomy as a framework. The study analyzed over 20,000 questions across various subjects, developing metrics like CogShift and category drift to measure cognitive depth. Findings indicate that specific prompting strategies can improve the quality and cognitive level of LLM-generated questions, suggesting potential for personalized learning systems. AI
IMPACT Highlights the need for cognitive-aware prompt design to improve LLMs for educational content creation.