Question Answering by Reasoning Across Documents with Graph Convolutional Networks

TL;DR

This paper presents a neural model that uses graph convolutional networks to perform multi-step reasoning across multiple documents for question answering, achieving state-of-the-art results on the WikiHop dataset.

Contribution

It introduces Entity-GCN, a scalable and compact graph-based neural model that reasons across documents using entity mentions and relations.

Findings

Achieves state-of-the-art results on WikiHop dataset.

Effectively models cross-document reasoning with GCNs.

Demonstrates scalability and compactness of the approach.

Abstract

Most research in reading comprehension has focused on answering questions based on individual documents or even single paragraphs. We introduce a neural model which integrates and reasons relying on information spread within documents and across multiple documents. We frame it as an inference problem on a graph. Mentions of entities are nodes of this graph while edges encode relations between different mentions (e.g., within- and cross-document co-reference). Graph convolutional networks (GCNs) are applied to these graphs and trained to perform multi-step reasoning. Our Entity-GCN method is scalable and compact, and it achieves state-of-the-art results on a multi-document question answering dataset, WikiHop (Welbl et al., 2018).