ROPES (Reasoning Over Paragraph Effects in Situations)

ROPES 是一个 QA 数据集，用于测试系统将文本段落中的知识应用于新情况的能力。系统被呈现一个包含因果关系或定性关系的背景段落、使用此背景的新情况以及需要在情境上下文中推理背景段落中的关系效果的问题。

ROPES is a question answering (QA) dataset developed to test a system's ability to apply knowledge from textual passages to novel scenarios. Each dataset instance includes a background passage containing causal or qualitative relationships, a novel scenario that leverages this background knowledge, and a question that requires the system to reason about the effects of the relationships in the background passage within the context of the scenario.

搜集汇总

背景与挑战

背景概述

ROPES是一个问答数据集，用于评估系统在给定背景段落和新情境下，推理因果关系或定性关系效果的能力。该数据集由艾伦人工智能研究所在2019年发布，旨在促进文本阅读理解中的情境推理任务。