Towards Interpretable Reasoning over Paragraph Effects in Situation

TL;DR

This paper introduces a sequential, interpretable neural network approach for reasoning over paragraph effects in situations, explicitly modeling each reasoning step to improve understanding and explainability in complex cause-effect tasks.

Contribution

It proposes a novel multi-module neural network that explicitly models each reasoning step, enhancing interpretability and performance over traditional black-box models.

Findings

Effective on ROPES dataset

Improves interpretability of reasoning process

Achieves better accuracy than baseline models

Abstract

We focus on the task of reasoning over paragraph effects in situation, which requires a model to understand the cause and effect described in a background paragraph, and apply the knowledge to a novel situation. Existing works ignore the complicated reasoning process and solve it with a one-step "black box" model. Inspired by human cognitive processes, in this paper we propose a sequential approach for this task which explicitly models each step of the reasoning process with neural network modules. In particular, five reasoning modules are designed and learned in an end-to-end manner, which leads to a more interpretable model. Experimental results on the ROPES dataset demonstrate the effectiveness and explainability of our proposed approach.