# SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators

arXiv: https://arxiv.org/abs/2502.06394 · HTML: https://mlsanity.com/paper/2502.06394/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 57 references that are in the map.
- First posted on arXiv 10 Feb 2025.

## Builds on (most cited first, 30 of 31)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Adafactor: Adaptive Learning Rates with Sublinear Memory Cost](https://mlsanity.com/paper/1804.04235/) (2018)
- [Language-agnostic BERT Sentence Embedding](https://mlsanity.com/paper/2007.01852/) (2020)
- [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)
- [ZeroGen: Efficient Zero-shot Learning via Dataset Generation](https://mlsanity.com/paper/2202.07922/) (2022)
- [A New Generation of Perspective API: Efficient Multilingual Character-level Transformers](https://mlsanity.com/paper/2202.11176/) (2022)
- [Reformulating Unsupervised Style Transfer as Paraphrase Generation](https://mlsanity.com/paper/2010.05700/) (2020)
- [Style Transfer in Text: Exploration and Evaluation](https://mlsanity.com/paper/1711.06861/) (2017)
- [Multilingual and Multi-Aspect Hate Speech Analysis](https://mlsanity.com/paper/1908.11049/) (2019)
- [Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier](https://mlsanity.com/paper/2412.04261/) (2024)
- [Style Transfer Through Back-Translation](https://mlsanity.com/paper/1804.09000/) (2018)
- [Text Classification via Large Language Models](https://mlsanity.com/paper/2305.08377/) (2023)
- [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)
- [Prompt-and-Rerank: A Method for Zero-Shot and Few-Shot Arbitrary Textual Style Transfer with Small Language Models](https://mlsanity.com/paper/2205.11503/) (2022)
- [Thank you BART! Rewarding Pre-Trained Models Improves Formality Style Transfer](https://mlsanity.com/paper/2105.06947/) (2021)
- [APPDIA: A Discourse-aware Transformer-based Style Transfer Model for Offensive Social Media Conversations](https://mlsanity.com/paper/2209.08207/) (2022)
- [TinyStyler: Efficient Few-Shot Text Style Transfer with Authorship Embeddings](https://mlsanity.com/paper/2406.15586/) (2024)
- [Exploring Methods for Cross-lingual Text Style Transfer: The Case of Text Detoxification](https://mlsanity.com/paper/2311.13937/) (2023)
- [MELA: Multilingual Evaluation of Linguistic Acceptability](https://mlsanity.com/paper/2311.09033/) (2023)
- [Multilingual and Explainable Text Detoxification with Parallel Corpora](https://mlsanity.com/paper/2412.11691/) (2024)
- [On the Role of Parallel Data in Cross-lingual Transfer Learning](https://mlsanity.com/paper/2212.10173/) (2022)
- [Authorship Style Transfer with Policy Optimization](https://mlsanity.com/paper/2403.08043/) (2024)
- [DetoxLLM: A Framework for Detoxification with Explanations](https://mlsanity.com/paper/2402.15951/) (2024)
- [Clandestino or Rifugiato? Anti-immigration Facebook Ad Targeting in Italy](https://mlsanity.com/paper/2103.09224/) (2021)

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

- [GemDetox at TextDetox CLEF 2025: Enhancing a Massively Multilingual Model for Text Detoxification on Low-resource Languages](https://mlsanity.com/paper/2510.01250/) (2025)
- [Text Detoxification in isiXhosa and Yor`ub'a: A Cross-Lingual Machine Learning Approach for Low-Resource African Languages](https://mlsanity.com/paper/2601.05624/) (2026)
- [A Survey of Toxicity Detection and Mitigation Strategies for Multilingual Language Models](https://mlsanity.com/paper/2606.25380/) (2026)
- [The Tatoxa System for Text Detoxification in Low-Resource Languages: The Case of Tatar](https://mlsanity.com/paper/2606.26015/) (2026)

## Abstract (arXiv)

Existing approaches to multilingual text detoxification are hampered by the scarcity of parallel multilingual datasets. In this work, we introduce a pipeline for the generation of multilingual parallel detoxification data. We also introduce SynthDetoxM, a manually collected and synthetically generated multilingual parallel text detoxification dataset comprising 16,000 high-quality detoxification sentence pairs across German, French, Spanish and Russian. The data was sourced from different toxicity evaluation datasets and then rewritten with nine modern open-source LLMs in few-shot setting. Our experiments demonstrate that models trained on the produced synthetic datasets have superior performance to those trained on the human-annotated MultiParaDetox dataset even in data limited setting. Models trained on SynthDetoxM outperform all evaluated LLMs in few-shot setting. We release our dataset and code to help further research in multilingual text detoxification.