A Parallel-Hierarchical Model for Machine Comprehension on Sparse Data

Abstract

Understanding unstructured text is a major goal within natural language processing. Comprehension tests pose questions based on short text passages to evaluate such understanding. In this work, we investigate machine comprehension on the challenging {\it MCTest} benchmark. Partly because of its limited size, prior work on {\it MCTest} has focused mainly on engineering better features. We tackle the dataset with a neural approach, harnessing simple neural networks arranged in a parallel hierarchy. The parallel hierarchy enables our model to compare the passage, question, and answer from a variety of trainable perspectives, as opposed to using a manually designed, rigid feature set. Perspectives range from the word level to sentence fragments to sequences of sentences; the networks operate only on word-embedding representations of text. When trained with a methodology designed to help cope with limited training data, our Parallel-Hierarchical model sets a new state of the art for {\it MCTest}, outperforming previous feature-engineered approaches slightly and previous neural approaches by a significant margin (over 15\% absolute).

9 pages, submitted to ACL

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Efficient Estimation of Word Representations in Vector Space
• Teaching Machines to Read and Comprehend
• Ask Me Anything: Dynamic Memory Networks for Natural Language Processing
• A Simple Way to Initialize Recurrent Networks of Rectified Linear Units
• The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations
• Learning Natural Language Inference with LSTM
• Towards Neural Network-based Reasoning
• Attention-Based Convolutional Neural Network for Machine Comprehension

Cited by in corpus (10)

• Embracing data abundance: BookTest Dataset for Reading Comprehension
• Option Comparison Network for Multiple-choice Reading Comprehension
• End-to-End Answer Chunk Extraction and Ranking for Reading Comprehension
• A Co-Matching Model for Multi-choice Reading Comprehension
• Supervised and Unsupervised Transfer Learning for Question Answering
• Hierarchical Question Answering for Long Documents
• Universal Natural Language Processing with Limited Annotations: Try Few-shot Textual Entailment as a Start
• Using Natural Language Relations between Answer Choices for Machine Comprehension
• Machine Comprehension Based on Learning to Rank
• A Base Camp for Scaling AI