Text Style Transfer
80 papers with code • 2 benchmarks • 6 datasets
Text Style Transfer is the task of controlling certain attributes of generated text. The state-of-the-art methods can be categorized into two main types which are used on parallel and non-parallel data. Methods on parallel data are typically supervised methods that use a neural sequence-to-sequence model with the encoder-decoder architecture. Methods on non-parallel data are usually unsupervised approaches using Disentanglement, Prototype Editing and Pseudo-Parallel Corpus Construction.
The popular benchmark for this task is the Yelp Review Dataset. Models are typically evaluated with the metrics of Sentiment Accuracy, BLEU, and PPL.
Benchmarks
These leaderboards are used to track progress in Text Style Transfer
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
Libraries
Use these libraries to find Text Style Transfer models and implementations
Yelp
Touchdown Dataset
TextSeg
StylePTB
XFORMAL
TextBox 2.0
Most implemented papers
Style Transfer from Non-Parallel Text by Cross-Alignment
shentianxiao/language-style-transfer
•
•
NeurIPS 2017
We demonstrate the effectiveness of this cross-alignment method on three tasks: sentiment modification, decipherment of word substitution ciphers, and recovery of word order.
A Probabilistic Formulation of Unsupervised Text Style Transfer
cindyxinyiwang/deep-latent-sequence-model
•
•
ICLR 2020
Across all style transfer tasks, our approach yields substantial gains over state-of-the-art non-generative baselines, including the state-of-the-art unsupervised machine translation techniques that our approach generalizes.
Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representation
fastnlp/style-transformer
•
•
ACL 2019
Disentangling the content and style in the latent space is prevalent in unpaired text style transfer.
Style Transfer Through Back-Translation
shrimai/Style-Transfer-Through-Back-Translation
•
•
ACL 2018
We first learn a latent representation of the input sentence which is grounded in a language translation model in order to better preserve the meaning of the sentence while reducing stylistic properties.
Disentangled Representation Learning for Non-Parallel Text Style Transfer
vineetjohn/linguistic-style-transfer
•
•
ACL 2019
This paper tackles the problem of disentangling the latent variables of style and content in language models.
Multiple-Attribute Text Style Transfer
martiansideofthemoon/style-transfer-paraphrase
•
•
The dominant approach to unsupervised "style transfer" in text is based on the idea of learning a latent representation, which is independent of the attributes specifying its "style".
IMaT: Unsupervised Text Attribute Transfer via Iterative Matching and Translation
Text attribute transfer aims to automatically rewrite sentences such that they possess certain linguistic attributes, while simultaneously preserving their semantic content.
Educating Text Autoencoders: Latent Representation Guidance via Denoising
shentianxiao/text-autoencoders
•
•
ICML 2020
We prove that this simple modification guides the latent space geometry of the resulting model by encouraging the encoder to map similar texts to similar latent representations.
Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters
astonzhang/Parameterization-of-Hypercomplex-Multiplications
•
•
Recent works have demonstrated reasonable success of representation learning in hypercomplex space.
IT5: Large-scale Text-to-text Pretraining for Italian Language Understanding and Generation
gsarti/it5
•
•
The T5 model and its unified text-to-text paradigm contributed in advancing the state-of-the-art for many natural language processing tasks.
