Natural Language Comprehension with the EpiReader
Adam Trischler adam.trischler、Zheng Ye jeff.ye、Xingdi Yuan eric.yuan \ANDKaheer Suleman Maluuba Research Montreal, Québec, Canada
Abstract
We present the EpiReader, a novel model for machine comprehension of text. Machine comprehension of unstructured, real-world text is a major research goal for natural language processing. Current tests of machine comprehension pose questions whose answers can be inferred from some supporting text, and evaluate a model’s response to the questions. The EpiReader is an end-to-end neural model comprising two components: the first component proposes a small set of candidate answers after comparing a question to its supporting text, and the second component formulates hypotheses using the proposed candidates and the question, then reranks the hypotheses based on their estimated concordance with the supporting text. We present experiments demonstrating that the EpiReader sets a new state-of-the-art on the CNN and Children’s Book Test machine comprehension benchmarks, outperforming previous neural models by a significant margin.
中文速览
机器阅读理解（machine comprehension）的难点在于：系统不仅要找到与问题相关的词，还要真正"理解"候选答案与原文之间的语义联系。EpiReader 把这个过程拆成两步：先用一个基于指针网络的"提取器"（Extractor）从原文中圈出少数几个最可能的候选答案，再用一个"推理器"（Reasoner）把每个候选词填回问题里构成完整假设，然后逐句比对假设与原文的文本蕴含关系（textual entailment），对候选答案重新打分排序。这种"提出假设→逐一验证"的思路来自古希腊哲学家伊壁鸠鲁的"多重解释原则"，也与人类阅读时边猜边验证的直觉相吻合。实验结果表明，EpiReader 在 CNN 新闻问答和儿童书籍测试（Children's Book Test）两个大规模基准上均刷新了当时的最优成绩，显著超越此前所有神经网络模型，证明把"候选筛选"与"语义推理"解耦并串联是提升机器阅读理解性能的有效途径。
原文 arXiv:1606.02270;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1606.02270v2