Speed Reading: Learning to Read ForBackward via Shuttle (2018-01-01T00:00:00.000000Z)

TL;DR

This work presents LSTM-Shuttle, which applies human speed reading techniques to natural language processing tasks for accurate and efficient comprehension and shows that it predicts both better and more quickly.

Abstract

We present LSTM-Shuttle, which applies human speed reading techniques to natural language processing tasks for accurate and efficient comprehension. In contrast to previous work, LSTM-Shuttle not only reads shuttling forward but also goes back. Shuttling forward enables high efficiency, and going backward gives the model a chance to recover lost information, ensuring better prediction. We evaluate LSTM-Shuttle on sentiment analysis, news classification, and cloze on IMDB, Rotten Tomatoes, AG, and Children’s Book Test datasets. We show that LSTM-Shuttle predicts both better and more quickly. To demonstrate how LSTM-Shuttle actually behaves, we also analyze the shuttling operation and present a case study.

Authors

Tsu-Jui Fu

3 papers

Wei-Yun Ma

4 papers

Speed Reading: Learning to Read ForBackward via Shuttle

TL;DR

Abstract

Authors

References28 items

A Simple and Effective Approach to the Story Cloze Test

Deep learning for sentiment analysis: A survey

Neural Speed Reading via Skim-RNN

Automatic differentiation in PyTorch

Skip RNN: Learning to Skip State Updates in Recurrent Neural Networks

Text Summarization Techniques: A Brief Survey

Deal or No Deal? End-to-End Learning of Negotiation Dialogues

Learning to Skim Text

Reading Wikipedia to Answer Open-Domain Questions

Coherent Dialogue with Attention-Based Language Models

ReasoNet: Learning to Stop Reading in Machine Comprehension

A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task

Named Entity Recognition with Bidirectional LSTM-CNNs

The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations

Part-of-Speech Tagging with Bidirectional Long Short-Term Memory Recurrent Neural Network

Neural Responding Machine for Short-Text Conversation

Adam: A Method for Stochastic Optimization

Neural Machine Translation by Jointly Learning to Align and Translate

Recurrent Models of Visual Attention

Distributed Representations of Sentences and Documents

Learning Word Vectors for Sentiment Analysis

Open Knowledge Extraction through Compositional Language Processing

Seeing Stars: Exploiting Class Relationships for Sentiment Categorization with Respect to Rating Scales

Policy Gradient Methods for Reinforcement Learning with Function Approximation

Long Short-Term Memory

Simple statistical gradient-following algorithms for connectionist reinforcement learning

Fast and Accurate Text Classification: Skimming, Rereading and Early Stopping

Convolutional neural net- works for sentence classification

Field of Study

Venue Information

Name

Type

URL

Alternate Names