MedGame uses LLMs to gamify medical education with interactive storytelling

Researchers have developed MedGame, a framework designed to enhance medical education by transforming static clinical cases into interactive, gamified storytelling experiences. This system utilizes a dual-engine approach, with a Medical Narrative Designer creating case-grounded storylines and a Story Director orchestrating multimodal plans. To evaluate its effectiveness, a benchmark called MedGame Bench, comprising 5,000 cases, was created for assessing Medical Narrative Generation and Story Direction. Initial experiments indicate that fine-tuning open-source large language models improves their performance on this benchmark, and a pilot study suggests students find MedGame more engaging than traditional text-based methods. AI

IMPACT This framework could significantly enhance engagement and learning outcomes in medical education by providing interactive, case-based learning experiences.

RANK_REASON The cluster describes a new research paper detailing a novel framework and benchmark for using LLMs in medical education. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• Hugging Face
• large-language models
• MedGame
• MedGame Bench
• Medical Narrative Generation
• Story Direction

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →