reddgr/nli-chatbot-prompt-categorization

该数据集包含带有自然语言推理（NLI）类别假设和标签的聊天机器人提示。其主要目的是对聊天机器人对话进行自然语言推理分类，这些对话通常与新闻、科学文章等常见NLI数据集中的文本模式和类别有很大不同。类别假设（语言、编码、角色扮演、科学等）被选为聊天机器人对话中最相关的，因为其语言上下文和主要使用场景通常与这些常见数据集中的文本模式和类别差异较大。

This dataset contains chatbot prompts annotated with natural language inference (NLI) category hypotheses and labels. The primary purpose is to perform natural language inference categorization of chatbot conversations. Category hypotheses (such as language, coding, role play, science, etc.) are chosen based on the context and main use cases of chatbot conversations, which typically differ significantly from the text patterns and categories found in the most popular NLI datasets sourced from news, scientific articles, and news publications.