FeedbackQA

该数据集是一个基于检索的问答数据集，包含了交互式的用户反馈。每个问题-答案对都被从“优秀”到“差”的不同等级进行评分，并附有详细说明回应优势和不足的自然语言解释。此外，该数据集还包括了对回应优势和不足的自然语言解释，以便更深入地理解答案的质量。其所涉及的任务是基于检索的问答。

This dataset is a retrieval-based question answering dataset that incorporates interactive user feedback. Each question-answer pair is rated on a scale ranging from "Excellent" to "Poor", accompanied by detailed natural language explanations elaborating on the strengths and weaknesses of the corresponding responses. Additionally, the dataset includes such natural language explanations of response strengths and weaknesses to enable a deeper understanding of answer quality. The core task involved in this dataset is retrieval-based question answering.