Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representation

Ning Dai, Jianze Liang, Xipeng Qiu, Xuanjing Huang
5/14/2019
cs.CL

Official code: FudanNLP/style-transformer (180 stars, last commit 2022-02-06). 1 more implementation below.

Abstract

Disentangling the content and style in the latent space is prevalent in unpaired text style transfer. However, two major issues exist in most of the current neural models. 1) It is difficult to completely strip the style information from the semantics for a sentence. 2) The recurrent neural network (RNN) based encoder and decoder, mediated by the latent representation, cannot well deal with the issue of the long-term dependency, resulting in poor preservation of non-stylistic semantic content. In this paper, we propose the Style Transformer, which makes no assumption about the latent representation of source sentence and equips the power of attention mechanism in Transformer to achieve better style transfer and better content preservation.

View on arXivView PDF

Code Implementations(2)

FudanNLP/style-transformerOfficial100%
18056Roff, PythonAug 16, 20194 years ago
FudanNLP/nlp-datasetOfficial100%

List some datasets in NLP field.

295Jun 20, 20195 years ago

Similar Papers

Disentangled Representation Learning for Non-Parallel Text Style Transfer

85% match1 repo1452018

Structured Content Preservation for Unsupervised Text Style Transfer

83% match1 repo332018

STEER: Unified Style Transfer with Expert Reinforcement

82% match1 repo92023

CLIPstyler: Image Style Transfer with a Single Text Condition

82% match1 repo3262021

A Closed-form Solution to Universal Style Transfer

81% match1 repo332019

Style Transfer from Non-Parallel Text by Cross-Alignment

81% match1 repo5582017

Cite this paper

@article{dai2019style,
  title  = {Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representation},
  author = {Ning Dai and Jianze Liang and Xipeng Qiu and Xuanjing Huang},
  year   = {2019},
  eprint = {1905.05621},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1905.05621}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1905.05621.svg)](https://codeofpaper.com/paper/1905.05621)

Discussion