DP-GAN: Diversity-Promoting Generative Adversarial Network for Generating Informative and Diversified Text

Abstract

Existing text generation methods tend to produce repeated and "boring" expressions. To tackle this problem, we propose a new text generation model, called Diversity-Promoting Generative Adversarial Network (DP-GAN). The proposed model assigns low reward for repeatedly generated text and high reward for "novel" and fluent text, encouraging the generator to produce diverse and informative text. Moreover, we propose a novel language-model based discriminator, which can better distinguish novel text from repeated text without the saturation problem compared with existing classifier-based discriminators. The experimental results on review generation and dialogue generation tasks demonstrate that our model can generate substantially more diverse and informative text than existing baselines. The code is available at https://github.com/lancopku/DPGAN

Accepted by EMNLP 2018

References in corpus (20)

• Neural Machine Translation by Jointly Learning to Align and Translate
• Sequence to Sequence Learning with Neural Networks
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
• Improved Techniques for Training GANs
• InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
• Energy-based Generative Adversarial Network
• Effective Approaches to Attention-based Neural Machine Translation
• BEGAN: Boundary Equilibrium Generative Adversarial Networks
• f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization
• How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
• An Actor-Critic Algorithm for Sequence Prediction
• A Hierarchical Neural Autoencoder for Paragraphs and Documents
• From Amateurs to Connoisseurs: Modeling the Evolution of User Expertise through Online Reviews
• Table-to-text Generation by Structure-aware Seq2seq Learning
• Generating Sentences by Editing Prototypes
• Global Encoding for Abstractive Summarization
• Autoencoder as Assistant Supervisor: Improving Text Representation for Chinese Social Media Text Summarization
• An Auto-Encoder Matching Model for Learning Utterance-Level Semantic Dependency in Dialogue Generation

Cited by in corpus (25)

• Synthesizing Tabular Data using Generative Adversarial Networks
• Language GANs Falling Short
• On Accurate Evaluation of GANs for Language Generation
• Neural Text Generation: Past, Present and Beyond
• ControlVAE: Controllable Variational Autoencoder
• Differentially Private Synthetic Data: Applied Evaluations and Enhancements
• Interpreting Deep Learning Models in Natural Language Processing: A Review
• Neural Language Generation: Formulation, Methods, and Evaluation
• Interpretable Directed Diversity: Leveraging Model Explanations for Iterative Crowd Ideation
• Privacy-Preserving Data Sharing in Agriculture: Enforcing Policy Rules for Secure and Confidential Data Synthesis
• OpenViDial: A Large-Scale, Open-Domain Dialogue Dataset with Visual Contexts
• Modeling Text-visual Mutual Dependency for Multi-modal Dialog Generation
• Bag-of-Words as Target for Neural Machine Translation
• A Sinogram Inpainting Method based on Generative Adversarial Network for Limited-angle Computed Tomography
• Regularizing Output Distribution of Abstractive Chinese Social Media Text Summarization for Improved Semantic Consistency
• OpenViDial 2.0: A Larger-Scale, Open-Domain Dialogue Generation Dataset with Visual Contexts
• Adversarial Attack and Defense of Structured Prediction Models
• Semi-supervised Conditional GAN for Simultaneous Generation and Detection of Phishing URLs: A Game theoretic Perspective
• Using GAN-based models to sentimental analysis on imbalanced datasets in education domain
• Meta-CoTGAN: A Meta Cooperative Training Paradigm for Improving Adversarial Text Generation
• ControlVAE: Tuning, Analytical Properties, and Performance Analysis
• EnsembleGAN: Adversarial Learning for Retrieval-Generation Ensemble Model on Short-Text Conversation
• HSCJN: A Holistic Semantic Constraint Joint Network for Diverse Response Generation
• Adversarial Stein Training for Graph Energy Models
• Positioning yourself in the maze of Neural Text Generation: A Task-Agnostic Survey