New LLM patient simulator models dynamic emotions for palliative care training

Researchers have developed EmoPatient, a novel patient simulator that uses large language models to generate dynamic emotional responses for palliative care communication training. Unlike static simulators, EmoPatient incorporates an "Emotion Director" agent to manage evolving emotional states and provide turn-level control signals for emotional intensity and regulatory stability. Evaluations in simulated physician-patient dialogues demonstrated improved emotional realism and robustness across different conversational personalities, suggesting that modeling emotional dynamics enhances the effectiveness of LLM-based training tools. AI

IMPACT Enhances realism in AI-driven medical training simulations, potentially improving clinician communication skills.

RANK_REASON Academic paper detailing a new LLM-based simulation tool. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• EmoPatient
• Emotion Director
• Gotit.pub
• Hugging Face
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• product

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