Existing question answering systems can only predict answers without explicit reasoning processes, which hinder their explainability and make us overestimate their ability of understanding and reasoning over natural language. In this work, we propose a novel task of reading comprehension, in which a model is required to provide final answers and reasoning processes. To this end, we introduce a formalism for reasoning over unstructured text, namely Text Reasoning Meaning Representation (TRMR). TRMR consists of three phrases, which is expressive enough to characterize the reasoning process to answer reading comprehension questions. We develop an annotation platform to facilitate TRMR's annotation, and release the R3 dataset, a \textbf{R}eading comprehension benchmark \textbf{R}equiring \textbf{R}easoning processes. R3 contains over 60K pairs of question-answer pairs and their TRMRs. Our dataset is available at: \url{http://anonymous}.

中文翻译：

---

R3：需要推理过程的阅读理解基准

现有的问答系统只能在没有明确推理过程的情况下预测答案，这阻碍了它们的可解释性，并使我们高估了它们对自然语言的理解和推理能力。在这项工作中，我们提出了一项新的阅读理解任务，其中需要一个模型来提供最终答案和推理过程。为此，我们引入了一种对非结构化文本进行推理的形式主义，即文本推理意义表示（TRMR）。TRMR 由三个短语组成，其表达能力足以表征回答阅读理解问题的推理过程。我们开发了一个注释平台来促进 TRMR 的注释，并发布 R3 数据集，一个 \textbf{R} 阅读理解基准 \textbf{R} 需要 \textbf{R} 推理过程。R3 包含超过 60K 对问答对及其 TRMR。我们的数据集位于：\url{http://anonymous}。