Multi-Perspective Context Matching for Machine Comprehension
Zhiguo Wang, Haitao Mi, Wael Hamza and Radu Florian IBM T.J. Watson Research Center 1101 Kitchawan Rd, Yorktown Heights, NY 10598
Abstract
Previous machine comprehension (MC) datasets are either too small to train end-to-end deep learning models, or not difficult enough to evaluate the ability of current MC techniques. The newly released SQuAD dataset alleviates these limitations, and gives us a chance to develop more realistic MC models. Based on this dataset, we propose a Multi-Perspective Context Matching (MPCM) model, which is an end-to-end system that directly predicts the answer beginning and ending points in a passage. Our model first adjusts each word-embedding vector in the passage by multiplying a relevancy weight computed against the question. Then, we encode the question and weighted passage by using bi-directional LSTMs. For each point in the passage, our model matches the context of this point against the encoded question from multiple perspectives and produces a matching vector. Given those matched vectors, we employ another bi-directional LSTM to aggregate all the information and predict the beginning and ending points. Experimental result on the test set of SQuAD shows that our model achieves a competitive result on the leaderboard.
中文速览
机器阅读理解（Machine Comprehension）长期面临两难困境：小数据集难以训练深度神经网络，大规模自动生成的完形填空数据集又太简单、不能真实反映模型能力。针对这一问题，研究者利用更贴近真实场景的SQuAD数据集，提出了一种端到端的多视角上下文匹配模型（Multi-Perspective Context Matching，MPCM），其核心思路是：一段文字越可能是答案，它的上下文就越应该与问题相似。模型先用问题相关度对段落词向量加权过滤，再通过双向LSTM编码，然后从全局匹配、最大池化匹配、均值池化匹配等多个视角将段落中每个位置的上下文与问题进行比较，最终独立预测答案的起始和结束位置。实验结果显示，单模型在SQuAD测试集上达到65.5 EM / 75.1 F1，集成5个模型后还能再提升约3个百分点，在当时的排行榜上取得了有竞争力的成绩；消融实验也证明模型各组件缺一不可，这为后续阅读理解研究提供了一种简洁而有效的多视角匹配框架。
原文 arXiv:1612.04211;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1612.04211v1