Evaluation of large language models in pediatric dentistry: a Bloom's taxonomy-based analysis.

Key Points

• The study found that large language models can accurately answer pediatric dentistry questions, demonstrating significant potential in medical education.
• Accuracy rates were specifically analyzed for ChatGPT-4.0, Claude 3.5 Sonnet, and DeepSeek R1, revealing varying performance levels among these models.
• Assessment utilized Bloom's taxonomy to evaluate justification quality, ensuring comprehensive analysis of cognitive skills.
• Findings support the integration of advanced AI tools in educational settings, highlighting their role in enhancing learning and assessment strategies.