A Hierarchical Reinforced Sequence Operation Method for Unsupervised Text Style Transfer

TL;DR

This paper introduces a hierarchical reinforcement learning approach called PTO for unsupervised text style transfer, improving interpretability, content preservation, and style-content trade-off handling.

Contribution

The paper proposes a novel hierarchical reinforced sequence operation method with a high-level position proposal and low-level sentence editing, enhancing unsupervised text style transfer.

Findings

Outperforms recent methods on style transfer datasets

Effectively balances content preservation and style control

Enables multi-step revision with a single-step trained model

Abstract

Unsupervised text style transfer aims to alter text styles while preserving the content, without aligned data for supervision. Existing seq2seq methods face three challenges: 1) the transfer is weakly interpretable, 2) generated outputs struggle in content preservation, and 3) the trade-off between content and style is intractable. To address these challenges, we propose a hierarchical reinforced sequence operation method, named Point-Then-Operate (PTO), which consists of a high-level agent that proposes operation positions and a low-level agent that alters the sentence. We provide comprehensive training objectives to control the fluency, style, and content of the outputs and a mask-based inference algorithm that allows for multi-step revision based on the single-step trained agents. Experimental results on two text style transfer datasets show that our method significantly outperforms recent methods and effectively addresses the aforementioned challenges.