# Evaluating Text Style Transfer: A Nine-Language Benchmark for Text Detoxification

arXiv: https://arxiv.org/abs/2507.15557 · HTML: https://mlsanity.com/paper/2507.15557/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 42 references that are in the map.
- First posted on arXiv 21 Jul 2025.

## Builds on (most cited first, 22 of 22)

- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Language-agnostic BERT Sentence Embedding](https://mlsanity.com/paper/2007.01852/) (2020)
- [LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods](https://mlsanity.com/paper/2412.05579/) (2024)
- [xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detection](https://mlsanity.com/paper/2310.10482/) (2023)
- [LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks](https://mlsanity.com/paper/2406.18403/) (2024)
- [Dear Sir or Madam, May I introduce the GYAFC Dataset: Corpus, Benchmarks and Metrics for Formality Style Transfer](https://mlsanity.com/paper/1803.06535/) (2018)
- [Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfer](https://mlsanity.com/paper/1804.06437/) (2018)
- [Automatically Neutralizing Subjective Bias in Text](https://mlsanity.com/paper/1911.09709/) (2019)
- [Fighting Offensive Language on Social Media with Unsupervised Text Style Transfer](https://mlsanity.com/paper/1805.07685/) (2018)
- [XFORMAL: A Benchmark for Multilingual Formality Style Transfer](https://mlsanity.com/paper/2104.04108/) (2021)
- [Text Detoxification using Large Pre-trained Neural Models](https://mlsanity.com/paper/2109.08914/) (2021)
- [CompassJudger-1: All-in-one Judge Model Helps Model Evaluation and Evolution](https://mlsanity.com/paper/2410.16256/) (2024)
- [Detoxifying Text with MaRCo: Controllable Revision with Experts and Anti-Experts](https://mlsanity.com/paper/2212.10543/) (2022)
- [On the Evaluation Metrics for Paraphrase Generation](https://mlsanity.com/paper/2202.08479/) (2022)
- [Evaluating prose style transfer with the Bible](https://mlsanity.com/paper/1711.04731/) (2017)
- [NLP for Counterspeech against Hate: A Survey and How-To Guide](https://mlsanity.com/paper/2403.20103/) (2024)
- [Exploring Methods for Cross-lingual Text Style Transfer: The Case of Text Detoxification](https://mlsanity.com/paper/2311.13937/) (2023)
- [Text Detoxification as Style Transfer in English and Hindi](https://mlsanity.com/paper/2402.07767/) (2024)
- [xCOMET-lite: Bridging the Gap Between Efficiency and Quality in Learned MT Evaluation Metrics](https://mlsanity.com/paper/2406.14553/) (2024)
- [SmurfCat at PAN 2024 TextDetox: Alignment of Multilingual Transformers for Text Detoxification](https://mlsanity.com/paper/2407.05449/) (2024)

## Built on by (most cited first, 4 of 4)

- [BhashaSutra: A Task-Centric Unified Survey of Indian NLP Datasets, Corpora, and Resources](https://mlsanity.com/paper/2604.18423/) (2026)
- [RT-SFT: Text Style Transfer from Non-Parallel Corpora by Roundtrip Translation](https://mlsanity.com/paper/2602.15013/) (2026)
- [The Tatoxa System for Text Detoxification in Low-Resource Languages: The Case of Tatar](https://mlsanity.com/paper/2606.26015/) (2026)
- [Fantastic Metrics and How to Find Them: Meta-Evaluating Metrics for Multi-lingual Style-Personalized Generation](https://mlsanity.com/paper/2610.19166/) (2026)

## Abstract (arXiv)

Despite notable advances in large language models (LLMs), reliable evaluation of text generation tasks such as text style transfer (TST) remains an open challenge. Existing research has shown that automatic metrics often correlate poorly with human judgments (Dementieva et al., 2024; Pauli et al., 2025), limiting our ability to assess model performance accurately. Furthermore, most prior work has focused primarily on English, while the evaluation of multilingual TST systems, particularly for text detoxification, remains largely underexplored. In this paper, we present the first comprehensive multilingual benchmarking study of evaluation metrics for text detoxification evaluation across nine languages: Arabic, Amharic, Chinese, English, German, Hindi, Russian, Spanish, and Ukrainian. Drawing inspiration from machine translation evaluation, we compare neural-based automatic metrics with LLM-as-a-judge approaches together with experiments on task-specific fine-tuned models. Our analysis reveals that the proposed metrics achieve significantly higher correlation with human judgments compared to baseline approaches. We also provide actionable insights and practical guidelines for building robust and reliable multilingual evaluation pipelines for text detoxification and related TST tasks.