# Style Obfuscation by Invariance

arXiv: https://arxiv.org/abs/1805.07143 · HTML: https://mlsanity.com/paper/1805.07143/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 May 2018.

## Built on by (most cited first, 10 of 10)

- [Civil Rephrases Of Toxic Texts With Self-Supervised Transformers](https://mlsanity.com/paper/2102.05456/) (2021)
- [Adversarial Stylometry in the Wild: Transferable Lexical Substitution Attacks on Author Profiling](https://mlsanity.com/paper/2101.11310/) (2021)
- [Style Pooling: Automatic Text Style Obfuscation for Improved Classification Fairness](https://mlsanity.com/paper/2109.04624/) (2021)
- [Keep It Private: Unsupervised Privatization of Online Text](https://mlsanity.com/paper/2405.10260/) (2024)
- [NAP^2: A Benchmark for Naturalness and Privacy-Preserving Text Rewriting by Learning from Human](https://mlsanity.com/paper/2406.03749/) (2024)
- [From Theories on Styles to their Transfer in Text: Bridging the Gap with a Hierarchical Survey](https://mlsanity.com/paper/2110.15871/) (2021)
- [Cyberbullying Classifiers are Sensitive to Model-Agnostic Perturbations](https://mlsanity.com/paper/2201.06384/) (2022)
- [Reproduction and Replication of an Adversarial Stylometry Experiment](https://mlsanity.com/paper/2208.07395/) (2022)
- [Defending Against Authorship Identification Attacks](https://mlsanity.com/paper/2310.01568/) (2023)
- [Protecting Anonymous Speech: A Generative Adversarial Network Methodology for Removing Stylistic Indicators in Text](https://mlsanity.com/paper/2110.09495/) (2021)

## Abstract (arXiv)

The task of obfuscating writing style using sequence models has previously been investigated under the framework of obfuscation-by-transfer, where the input text is explicitly rewritten in another style. These approaches also often lead to major alterations to the semantic content of the input. In this work, we propose obfuscation-by-invariance, and investigate to what extent models trained to be explicitly style-invariant preserve semantics. We evaluate our architectures on parallel and non-parallel corpora, and compare automatic and human evaluations on the obfuscated sentences. Our experiments show that style classifier performance can be reduced to chance level, whilst the automatic evaluation of the output is seemingly equal to models applying style-transfer. However, based on human evaluation we demonstrate a trade-off between the level of obfuscation and the observed quality of the output in terms of meaning preservation and grammaticality.