Joint Copying and Restricted Generation for Paraphrase

Abstract

Many natural language generation tasks, such as abstractive summarization and text simplification, are paraphrase-orientated. In these tasks, copying and rewriting are two main writing modes. Most previous sequence-to-sequence (Seq2Seq) models use a single decoder and neglect this fact. In this paper, we develop a novel Seq2Seq model to fuse a copying decoder and a restricted generative decoder. The copying decoder finds the position to be copied based on a typical attention model. The generative decoder produces words limited in the source-specific vocabulary. To combine the two decoders and determine the final output, we develop a predictor to predict the mode of copying or rewriting. This predictor can be guided by the actual writing mode in the training data. We conduct extensive experiments on two different paraphrase datasets. The result shows that our model outperforms the state-of-the-art approaches in terms of both informativeness and language quality.

7 pages, 1 figure, AAAI-17

Cited by in corpus (17)

• Unsupervised Paraphrase Generation using Pre-trained Language Models
• Paraphrase Generation with Deep Reinforcement Learning
• Coreferential Reasoning Learning for Language Representation
• Zero-Shot Paraphrase Generation with Multilingual Language Models
• Paraphrase Augmented Task-Oriented Dialog Generation
• Unsupervised Contextual Paraphrase Generation using Lexical Control and Reinforcement Learning
• Towards Diverse Paraphrase Generation Using Multi-Class Wasserstein GAN
• LAMPAT: Low-Rank Adaption for Multilingual Paraphrasing Using Adversarial Training
• Generating Syntactically Controlled Paraphrases without Using Annotated Parallel Pairs
• Decomposable Neural Paraphrase Generation
• PowerTransformer: Unsupervised Controllable Revision for Biased Language Correction
• Paraphrase Generation as Unsupervised Machine Translation
• ReDecode Framework for Iterative Improvement in Paraphrase Generation
• Towards Document-Level Paraphrase Generation with Sentence Rewriting and Reordering
• Unsupervised Paraphrasing with Pretrained Language Models
• ConRPG: Paraphrase Generation using Contexts as Regularizer
• Improving Scientific Article Visibility by Neural Title Simplification