TruthGen

该数据集为评估大型语言模型响应的真实性提供了一个基准。其任务在于检测大型语言模型响应中的真实性。通过对该数据集的分析，研究人员能够对大型语言模型在生成回答时保持真实性的能力进行客观评价和比较。

This dataset provides a benchmark for evaluating the factuality of responses generated by large language models (LLMs). Its core task is to detect the factuality of LLM-generated responses. Through analysis of this dataset, researchers can conduct objective evaluations and comparative assessments of the ability of LLMs to uphold factuality during the response generation process.