# Generative Sentiment Transfer via Adaptive Masking

## TL;DR

This paper introduces AM-ST, a novel adaptive masking model for sentiment transfer that learns to separate sentiment from content using attention mechanisms, improving transfer quality and semantic preservation.

## Contribution

It proposes a learnable masking approach and a sentiment-aware language model to enhance sentiment transfer performance over traditional rule-based methods.

## Key findings

- AM-ST outperforms existing methods on benchmark datasets.
- Adaptive masks improve sentiment transfer accuracy.
- Incorporating sentiment in language modeling captures multi-grained semantics.

## Abstract

Sentiment transfer aims at revising the input text to satisfy a given sentiment polarity while retaining the original semantic content. The nucleus of sentiment transfer lies in precisely separating the sentiment information from the content information. Existing explicit approaches generally identify and mask sentiment tokens simply based on prior linguistic knowledge and manually-defined rules, leading to low generality and undesirable transfer performance. In this paper, we view the positions to be masked as the learnable parameters, and further propose a novel AM-ST model to learn adaptive task-relevant masks based on the attention mechanism. Moreover, a sentiment-aware masked language model is further proposed to fill in the blanks in the masked positions by incorporating both context and sentiment polarity to capture the multi-grained semantics comprehensively. AM-ST is thoroughly evaluated on two popular datasets, and the experimental results demonstrate the superiority of our proposal.