Knowledgeable Reader: Enhancing Cloze-Style Reading Comprehension with External Commonsense Knowledge

TL;DR

This paper presents a neural reading comprehension model that enhances cloze-style question answering by integrating external commonsense knowledge through a key-value memory, improving accuracy and interpretability.

Contribution

It introduces a novel method for incorporating external knowledge into reading comprehension models using a key-value memory, outperforming strong baselines.

Findings

Improved accuracy on a challenging dataset.

Model can provide evidence of used background knowledge.

Outperforms more complex models on key metrics.

Abstract

We introduce a neural reading comprehension model that integrates external commonsense knowledge, encoded as a key-value memory, in a cloze-style setting. Instead of relying only on document-to-question interaction or discrete features as in prior work, our model attends to relevant external knowledge and combines this knowledge with the context representation before inferring the answer. This allows the model to attract and imply knowledge from an external knowledge source that is not explicitly stated in the text, but that is relevant for inferring the answer. Our model improves results over a very strong baseline on a hard Common Nouns dataset, making it a strong competitor of much more complex models. By including knowledge explicitly, our model can also provide evidence about the background knowledge used in the RC process.