Machine Comprehension Using Match-LSTM and Answer Pointer

TL;DR

This paper introduces an end-to-end neural network architecture combining Match-LSTM and Pointer Net for machine comprehension, effectively answering questions from the SQuAD dataset with improved accuracy over previous methods.

Contribution

It presents a novel neural model that integrates Match-LSTM and Pointer Net for question answering, advancing the state-of-the-art on SQuAD.

Findings

Models outperform previous logistic regression approaches

Both proposed models significantly improve accuracy

Effective handling of variable answer lengths

Abstract

Machine comprehension of text is an important problem in natural language processing. A recently released dataset, the Stanford Question Answering Dataset (SQuAD), offers a large number of real questions and their answers created by humans through crowdsourcing. SQuAD provides a challenging testbed for evaluating machine comprehension algorithms, partly because compared with previous datasets, in SQuAD the answers do not come from a small set of candidate answers and they have variable lengths. We propose an end-to-end neural architecture for the task. The architecture is based on match-LSTM, a model we proposed previously for textual entailment, and Pointer Net, a sequence-to-sequence model proposed by Vinyals et al.(2015) to constrain the output tokens to be from the input sequences. We propose two ways of using Pointer Net for our task. Our experiments show that both of our two models substantially outperform the best results obtained by Rajpurkar et al.(2016) using logistic regression and manually crafted features.