Sentence Simplification with Deep Reinforcement Learning

Abstract

Sentence simplification aims to make sentences easier to read and understand. Most recent approaches draw on insights from machine translation to learn simplification rewrites from monolingual corpora of complex and simple sentences. We address the simplification problem with an encoder-decoder model coupled with a deep reinforcement learning framework. Our model, which we call {\sc Dress} (as shorthand for {\bf D}eep {\bf RE}inforcement {\bf S}entence {\bf S}implification), explores the space of possible simplifications while learning to optimize a reward function that encourages outputs which are simple, fluent, and preserve the meaning of the input. Experiments on three datasets demonstrate that our model outperforms competitive simplification systems.

to appear in EMNLP 2017

References in corpus (3)

• Recurrent Neural Network Regularization
• Deep Reinforcement Learning for Dialogue Generation
• Improving Information Extraction by Acquiring External Evidence with Reinforcement Learning

Cited by in corpus (8)

• Reinforced Self-Attention Network: a Hybrid of Hard and Soft Attention for Sequence Modeling
• Integrating Transformer and Paraphrase Rules for Sentence Simplification
• Dynamic Multi-Level Multi-Task Learning for Sentence Simplification
• Text as Environment: A Deep Reinforcement Learning Text Readability Assessment Model
• Improving Neural Text Simplification Model with Simplified Corpora
• SHAPED: Shared-Private Encoder-Decoder for Text Style Adaptation
• Controlling Text Complexity in Neural Machine Translation
• Learning How to Self-Learn: Enhancing Self-Training Using Neural Reinforcement Learning