LLMs Can Simulate Standardized Patients via Agent Coevolution

TL;DR

EvoPatient introduces an agent coevolution framework where simulated patient and doctor agents learn through multi-turn dialogues, enhancing medical training by improving response quality and requirement alignment with efficient resource use.

Contribution

This work presents a novel unsupervised coevolution approach for simulated patient agents, enabling more realistic and effective medical training scenarios compared to prior methods.

Findings

Over 10% improvement in requirement alignment over existing methods

Achieved optimal resource consumption after 200 cases in 10 hours

Demonstrated strong generalizability across various medical cases

Abstract

Training medical personnel using standardized patients (SPs) remains a complex challenge, requiring extensive domain expertise and role-specific practice. Previous research on Large Language Model (LLM)-based SPs mostly focuses on improving data retrieval accuracy or adjusting prompts through human feedback. However, this focus has overlooked the critical need for patient agents to learn a standardized presentation pattern that transforms data into human-like patient responses through unsupervised simulations. To address this gap, we propose EvoPatient, a novel simulated patient framework in which a patient agent and doctor agents simulate the diagnostic process through multi-turn dialogues, simultaneously gathering experience to improve the quality of both questions and answers, ultimately enabling human doctor training. Extensive experiments on various cases demonstrate that, by providing only overall SP requirements, our framework improves over existing reasoning methods by more than 10\% in requirement alignment and better human preference, while achieving an optimal balance of resource consumption after evolving over 200 cases for 10 hours, with excellent generalizability. Our system will be available at https://github.com/ZJUMAI/EvoPatient.