非任务导向对话系统中文语料库

非任务导向对话系统中文语料库是由腾讯AI实验室创建的大规模语料库，包含超过27K独特提示和82K响应，数据来源于社交媒体。该数据集通过定义一个5级评分方案进行人工标注，旨在提高对话系统的响应选择质量。数据集的应用领域主要在于训练和评估对话系统，解决对话系统中响应质量不一的问题。

The Chinese non-task-oriented dialogue system corpus is a large-scale corpus created by Tencent AI Lab. It contains over 27K unique prompts and 82K responses, with data sourced from social media. This corpus was manually annotated via a predefined 5-level scoring scheme, aiming to improve the quality of response selection for dialogue systems. Its main applications are training and evaluating dialogue systems, to address the issue of inconsistent response quality in such systems.