Unpaired Sentiment-to-Sentiment Translation: A Cycled Reinforcement Learning Approach

Abstract

The goal of sentiment-to-sentiment "translation" is to change the underlying sentiment of a sentence while keeping its content. The main challenge is the lack of parallel data. To solve this problem, we propose a cycled reinforcement learning method that enables training on unpaired data by collaboration between a neutralization module and an emotionalization module. We evaluate our approach on two review datasets, Yelp and Amazon. Experimental results show that our approach significantly outperforms the state-of-the-art systems. Especially, the proposed method substantially improves the content preservation performance. The BLEU score is improved from 1.64 to 22.46 and from 0.56 to 14.06 on the two datasets, respectively.

arXiv commentAccepted by ACL 2018

References in corpus (5)

• Sequence to Sequence Learning with Neural Networks
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Dual Learning for Machine Translation
• Demystifying Neural Style Transfer
• Decoding-History-Based Adaptive Control of Attention for Neural Machine Translation

Cited by in corpus (26)

• Style Transfer as Unsupervised Machine Translation
• Multiple-Attribute Text Style Transfer
• Neural Language Generation: Formulation, Methods, and Evaluation
• Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory
• Conditional BERT Contextual Augmentation
• Decomposing Textual Information For Style Transfer
• From Theories on Styles to their Transfer in Text: Bridging the Gap with a Hierarchical Survey
• Grammatical Error Correction and Style Transfer via Zero-shot Monolingual Translation
• The Daunting Task of Real-World Textual Style Transfer Auto-Evaluation
• Cycle-Consistent Adversarial Autoencoders for Unsupervised Text Style Transfer
• XL-Editor: Post-editing Sentences with XLNet
• Hooks in the Headline: Learning to Generate Headlines with Controlled Styles
• Contextual Text Style Transfer
• ST${}^2$: Small-data Text Style Transfer via Multi-task Meta-Learning
• Multi-type Disentanglement without Adversarial Training
• Enhancing Content Preservation in Text Style Transfer Using Reverse Attention and Conditional Layer Normalization
• TextSETTR: Few-Shot Text Style Extraction and Tunable Targeted Restyling
• XFORMAL: A Benchmark for Multilingual Formality Style Transfer
• Opinion-aware Answer Generation for Review-driven Question Answering in E-Commerce
• DGST: a Dual-Generator Network for Text Style Transfer
• GTAE: Graph-Transformer based Auto-Encoders for Linguistic-Constrained Text Style Transfer
• Imbalanced Sentiment Classification Enhanced with Discourse Marker
• Learning to Generate Multiple Style Transfer Outputs for an Input Sentence
• Massive Styles Transfer with Limited Labeled Data
• Stylistic Retrieval-based Dialogue System with Unparallel Training Data
• SE-DAE: Style-Enhanced Denoising Auto-Encoder for Unsupervised Text Style Transfer