Neural Paraphrase Generation with Stacked Residual LSTM Networks

Abstract

In this paper, we propose a novel neural approach for paraphrase generation. Conventional para- phrase generation methods either leverage hand-written rules and thesauri-based alignments, or use statistical machine learning principles. To the best of our knowledge, this work is the first to explore deep learning models for paraphrase generation. Our primary contribution is a stacked residual LSTM network, where we add residual connections between LSTM layers. This allows for efficient training of deep LSTMs. We evaluate our model and other state-of-the-art deep learning models on three different datasets: PPDB, WikiAnswers and MSCOCO. Evaluation results demonstrate that our model outperforms sequence to sequence, attention-based and bi- directional LSTM models on BLEU, METEOR, TER and an embedding-based sentence similarity metric.

arXiv commentCOLING 2016

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Sequence to Sequence Learning with Neural Networks
• Skip-Thought Vectors
• End-To-End Memory Networks
• How to Construct Deep Recurrent Neural Networks
• Show and Tell: A Neural Image Caption Generator
• Generating Natural Language Inference Chains

Cited by in corpus (41)

• GODIVA: Generating Open-DomaIn Videos from nAtural Descriptions
• Understanding and Mitigating the Security Risks of Voice-Controlled Third-Party Skills on Amazon Alexa and Google Home
• Unsupervised Paraphrase Generation using Pre-trained Language Models
• Long-term Pedestrian Trajectory Prediction using Mutable Intention Filter and Warp LSTM
• Evaluating prose style transfer with the Bible
• Simulated annealing for optimization of graphs and sequences
• Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites
• Improving Natural Language Processing Tasks with Human Gaze-Guided Neural Attention
• Style-transfer and Paraphrase: Looking for a Sensible Semantic Similarity Metric
• How Large Language Models are Transforming Machine-Paraphrased Plagiarism
• Learning Semantic Sentence Embeddings using Sequential Pair-wise Discriminator
• Paraphrase Generation with Deep Reinforcement Learning
• Pushing Paraphrase Away from Original Sentence: A Multi-Round Paraphrase Generation Approach
• Zero-Shot Paraphrase Generation with Multilingual Language Models
• An Empirical Comparison on Imitation Learning and Reinforcement Learning for Paraphrase Generation
• COCO_TS Dataset: Pixel-level Annotations Based on Weak Supervision for Scene Text Segmentation
• Unsupervised Contextual Paraphrase Generation using Lexical Control and Reinforcement Learning
• Neural Poetry: Learning to Generate Poems using Syllables
• From Solving a Problem Boldly to Cutting the Gordian Knot: Idiomatic Text Generation
• Unsupervised Paraphrasing by Simulated Annealing
• Controllable Paraphrase Generation with a Syntactic Exemplar
• AutoQA: From Databases To QA Semantic Parsers With Only Synthetic Training Data
• Weakly Supervised Reasoning by Neuro-Symbolic Approaches
• Generating Syntactically Controlled Paraphrases without Using Annotated Parallel Pairs
• Syntax-Infused Variational Autoencoder for Text Generation
• Diversifying Task-oriented Dialogue Response Generation with Prototype Guided Paraphrasing
• PowerTransformer: Unsupervised Controllable Revision for Biased Language Correction
• Measuring Semantic Abstraction of Multilingual NMT with Paraphrase Recognition and Generation Tasks
• Polly Want a Cracker: Analyzing Performance of Parroting on Paraphrase Generation Datasets
• Paraphrase Generation as Unsupervised Machine Translation
• Exemplar-Controllable Paraphrasing and Translation using Bitext
• ReDecode Framework for Iterative Improvement in Paraphrase Generation
• Towards Document-Level Paraphrase Generation with Sentence Rewriting and Reordering
• MAPGN: MAsked Pointer-Generator Network for sequence-to-sequence pre-training
• Contrastive Representation Learning for Exemplar-Guided Paraphrase Generation
• Unsupervised Paraphrasing with Pretrained Language Models
• Paraphrase Thought: Sentence Embedding Module Imitating Human Language Recognition
• Improving the Diversity of Unsupervised Paraphrasing with Embedding Outputs
• Delexicalized Paraphrase Generation
• Modality Compensation Network: Cross-Modal Adaptation for Action Recognition
• Learning to Selectively Learn for Weakly-supervised Paraphrase Generation