Patient Simulator Dataset

该数据集包含了患者模拟器与大型语言模型（LLMs）之间的互动，旨在评估这些模型在诊断能力方面的表现。具体来说，数据集中包含了患者模拟器的详尽回应，专为评估大型语言模型在医疗诊断场景下的性能而设计。数据集的规模如下：训练集包含1890个样本，验证集包含300个样本，测试集包含500个样本。该任务旨在自动评估诊断对话。

This dataset contains interactions between patient simulators and Large Language Models (LLMs), with the goal of evaluating these models' diagnostic capabilities. Specifically, it includes detailed responses from patient simulators, which are specially designed to assess the performance of LLMs in medical diagnostic scenarios. The dataset has the following scale: the training set consists of 1,890 samples, the validation set has 300 samples, and the test set contains 500 samples. This task aims to automatically evaluate diagnostic conversations.