KG^2: Learning to Reason Science Exam Questions with Contextual Knowledge Graph Embeddings

TL;DR

This paper introduces a novel neural reasoning framework using contextual knowledge graph embeddings to improve science exam question answering on the challenging ARC dataset.

Contribution

It proposes a new method that constructs and reasons with knowledge graphs for questions and supporting sentences, enhancing reasoning capabilities.

Findings

Outperforms previous state-of-the-art QA systems on ARC Challenge Set

Demonstrates effective reasoning with knowledge graph embeddings

Addresses complex science question answering challenges

Abstract

The AI2 Reasoning Challenge (ARC), a new benchmark dataset for question answering (QA) has been recently released. ARC only contains natural science questions authored for human exams, which are hard to answer and require advanced logic reasoning. On the ARC Challenge Set, existing state-of-the-art QA systems fail to significantly outperform random baseline, reflecting the difficult nature of this task. In this paper, we propose a novel framework for answering science exam questions, which mimics human solving process in an open-book exam. To address the reasoning challenge, we construct contextual knowledge graphs respectively for the question itself and supporting sentences. Our model learns to reason with neural embeddings of both knowledge graphs. Experiments on the ARC Challenge Set show that our model outperforms the previous state-of-the-art QA systems.