Knowledgeable Reader: Enhancing Cloze-Style Reading Comprehension with External Commonsense Knowledge
Todor Mihaylov Anette Frank Research Training Group AIPHES Department of Computational Linguistics, Heidelberg University Heidelberg, Germany
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.
中文速览
用神经网络做完形填空式阅读理解时，模型往往只能靠文章本身的上下文来猜答案，遇到需要背景常识才能推断的题目就力不从心。这项研究提出了一个叫 KnReader（Knowledgeable Reader）的模型，把外部常识知识库（ConceptNet/WordNet）中检索到的相关知识三元组编码进一个键值记忆（key-value memory），让模型在理解文章每个词的时候都能"查阅"这份知识，并把知识表示直接融合进上下文向量，再去预测答案。在儿童书籍测试（Children's Book Test, CBT）数据集的普通名词子集上，该方法在强基线 Attention Sum Reader 基础上取得了显著提升，与结构复杂得多的多跳模型相比也不落下风。这项工作的价值在于：它证明了把外部常识显式注入简单模型就能弥补纯文本理解的短板，同时还能追溯模型到底用了哪条知识来推断答案，增强了可解释性。
原文 arXiv:1805.07858;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1805.07858v1