Query-Reduction Networks for Question Answering

Abstract

In this paper, we study the problem of question answering when reasoning over multiple facts is required. We propose Query-Reduction Network (QRN), a variant of Recurrent Neural Network (RNN) that effectively handles both short-term (local) and long-term (global) sequential dependencies to reason over multiple facts. QRN considers the context sentences as a sequence of state-changing triggers, and reduces the original query to a more informed query as it observes each trigger (context sentence) through time. Our experiments show that QRN produces the state-of-the-art results in bAbI QA and dialog tasks, and in a real goal-oriented dialog dataset. In addition, QRN formulation allows parallelization on RNN's time axis, saving an order of magnitude in time complexity for training and inference.

Published as a conference paper at ICLR 2017. Title of the paper has changed from "Query-Regression Networks for Machine Comprehension"

References in corpus (5)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• SQuAD: 100,000+ Questions for Machine Comprehension of Text
• Dynamic Memory Networks for Visual and Textual Question Answering
• Learning End-to-End Goal-Oriented Dialog
• Gated End-to-End Memory Networks

Cited by in corpus (25)

• Regularizing and Optimizing LSTM Language Models
• An End-to-End Trainable Neural Network Model with Belief Tracking for Task-Oriented Dialog
• Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approaches
• Revisiting Activation Regularization for Language RNNs
• Linguistic Knowledge as Memory for Recurrent Neural Networks
• Simulating Action Dynamics with Neural Process Networks
• Dialogue Learning with Human Teaching and Feedback in End-to-End Trainable Task-Oriented Dialogue Systems
• Weaver: Deep Co-Encoding of Questions and Documents for Machine Reading
• CmnRec: Sequential Recommendations with Chunk-accelerated Memory Network
• What Happened? Leveraging VerbNet to Predict the Effects of Actions in Procedural Text
• Sentiment Adaptive End-to-End Dialog Systems
• Procedural Reading Comprehension with Attribute-Aware Context Flow
• Hierarchical Question Answering for Long Documents
• A Heterogeneous Graph with Factual, Temporal and Logical Knowledge for Question Answering Over Dynamic Contexts
• Knowledge-incorporating ESIM models for Response Selection in Retrieval-based Dialog Systems
• Disentangling Language and Knowledge in Task-Oriented Dialogs
• NLProlog: Reasoning with Weak Unification for Question Answering in Natural Language
• Hybrid Code Networks using a convolutional neural network as an input layer achieves higher turn accuracy
• A Dataset for Building Code-Mixed Goal Oriented Conversation Systems
• Everything Happens for a Reason: Discovering the Purpose of Actions in Procedural Text
• Generative Question Refinement with Deep Reinforcement Learning in Retrieval-based QA System
• Effective Use of Transformer Networks for Entity Tracking
• Scaffolding Networks: Incremental Learning and Teaching Through Questioning
• Learning Invariants through Soft Unification
• Learning End-to-End Goal-Oriented Dialog with Multiple Answers