e-CARE: a New Dataset for Exploring Explainable Causal Reasoning

TL;DR

This paper introduces e-CARE, a new dataset with human-annotated causal reasoning questions and explanations, aiming to improve explainable causal reasoning in NLP, highlighting the challenge for current models and the benefit of explanations.

Contribution

The paper presents e-CARE, the first large-scale dataset with natural language explanations for causal reasoning, facilitating research in explainable NLP models.

Findings

Generating explanations remains challenging for state-of-the-art models.

Explanation data improves causal reasoning accuracy and stability.

e-CARE contains over 21,000 annotated causal questions.

Abstract

Understanding causality has vital importance for various Natural Language Processing (NLP) applications. Beyond the labeled instances, conceptual explanations of the causality can provide deep understanding of the causal facts to facilitate the causal reasoning process. However, such explanation information still remains absent in existing causal reasoning resources. In this paper, we fill this gap by presenting a human-annotated explainable CAusal REasoning dataset (e-CARE), which contains over 21K causal reasoning questions, together with natural language formed explanations of the causal questions. Experimental results show that generating valid explanations for causal facts still remains especially challenging for the state-of-the-art models, and the explanation information can be helpful for promoting the accuracy and stability of causal reasoning models.