Multiple-Attribute Text Style Transfer

Abstract

The dominant approach to unsupervised "style transfer" in text is based on the idea of learning a latent representation, which is independent of the attributes specifying its "style". In this paper, we show that this condition is not necessary and is not always met in practice, even with domain adversarial training that explicitly aims at learning such disentangled representations. We thus propose a new model that controls several factors of variation in textual data where this condition on disentanglement is replaced with a simpler mechanism based on back-translation. Our method allows control over multiple attributes, like gender, sentiment, product type, etc., and a more fine-grained control on the trade-off between content preservation and change of style with a pooling operator in the latent space. Our experiments demonstrate that the fully entangled model produces better generations, even when tested on new and more challenging benchmarks comprising reviews with multiple sentences and multiple attributes.

References in corpus (20)

• Adam: A Method for Stochastic Optimization
• Conditional Generative Adversarial Nets
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• Deep Convolutional Inverse Graphics Network
• Style Transfer from Non-Parallel Text by Cross-Alignment
• Dual Learning for Machine Translation
• Unsupervised Cross-Domain Image Generation
• Weakly-supervised Disentangling with Recurrent Transformations for 3D View Synthesis
• Word Translation Without Parallel Data
• Phrase-Based & Neural Unsupervised Machine Translation
• Unsupervised Machine Translation Using Monolingual Corpora Only
• Toward Controlled Generation of Text
• Style Transfer as Unsupervised Machine Translation
• Understanding Back-Translation at Scale
• Unpaired Sentiment-to-Sentiment Translation: A Cycled Reinforcement Learning Approach
• Unsupervised Text Style Transfer using Language Models as Discriminators
• Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfer
• Style Transfer in Text: Exploration and Evaluation
• Language Style Transfer from Sentences with Arbitrary Unknown Styles
• SHAPED: Shared-Private Encoder-Decoder for Text Style Adaptation

Cited by in corpus (13)

• Unsupervised Question Answering by Cloze Translation
• Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites
• Style-transfer and Paraphrase: Looking for a Sensible Semantic Similarity Metric
• Educating Text Autoencoders: Latent Representation Guidance via Denoising
• Decomposing Textual Information For Style Transfer
• Improving GAN Training with Probability Ratio Clipping and Sample Reweighting
• Domain Adaptive Text Style Transfer
• From Solving a Problem Boldly to Cutting the Gordian Knot: Idiomatic Text Generation
• Rethinking Text Attribute Transfer: A Lexical Analysis
• Contextual Text Style Transfer
• Neural Data-to-Text Generation with LM-based Text Augmentation
• Investigation of Sentiment Controllable Chatbot
• Learning to Generate Multiple Style Transfer Outputs for an Input Sentence