Balancing the Style-Content Trade-Off in Sentiment Transfer Using Polarity-Aware Denoising
This work addresses the style-content trade-off in sentiment transfer for natural language processing applications, representing an incremental improvement over existing methods.
The paper tackled the problem of insufficient content preservation in text sentiment transfer by proposing a polarity-aware denoising model, which outperformed state-of-the-art baselines in content preservation while maintaining competitive style transfer accuracy and fluency.
Text sentiment transfer aims to flip the sentiment polarity of a sentence (positive to negative or vice versa) while preserving its sentiment-independent content. Although current models show good results at changing the sentiment, content preservation in transferred sentences is insufficient. In this paper, we present a sentiment transfer model based on polarity-aware denoising, which accurately controls the sentiment attributes in generated text, preserving the content to a great extent and helping to balance the style-content trade-off. Our proposed model is structured around two key stages in the sentiment transfer process: better representation learning using a shared encoder and sentiment-controlled generation using separate sentiment-specific decoders. Empirical results show that our methods outperforms state-of-the-art baselines in terms of content preservation while staying competitive in terms of style transfer accuracy and fluency.