# Translate, then Detect: Leveraging Machine Translation for Cross-Lingual Toxicity Classification

arXiv: https://arxiv.org/abs/2509.14493 · HTML: https://mlsanity.com/paper/2509.14493/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 53 references that are in the map.
- First posted on arXiv 17 Sep 2025.

## Builds on (most cited first, 27 of 27)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfer](https://mlsanity.com/paper/2005.00052/) (2020)
- [Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond](https://mlsanity.com/paper/1812.10464/) (2018)
- [HateCheck: Functional Tests for Hate Speech Detection Models](https://mlsanity.com/paper/2012.15606/) (2020)
- [Exploiting Similarities among Languages for Machine Translation](https://mlsanity.com/paper/1309.4168/) (2013)
- [FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language](https://mlsanity.com/paper/2506.20920/) (2025)
- [Understanding Abuse: A Typology of Abusive Language Detection Subtasks](https://mlsanity.com/paper/1705.09899/) (2017)
- [Adaptation of Deep Bidirectional Multilingual Transformers for Russian Language](https://mlsanity.com/paper/1905.07213/) (2019)
- [Translation Artifacts in Cross-lingual Transfer Learning](https://mlsanity.com/paper/2004.04721/) (2020)
- [Do Multilingual Language Models Think Better in English?](https://mlsanity.com/paper/2308.01223/) (2023)
- [Scaling up COMETKIWI: Unbabel-IST 2023 Submission for the Quality Estimation Shared Task](https://mlsanity.com/paper/2309.11925/) (2023)
- [Toxic Language Detection in Social Media for Brazilian Portuguese: New Dataset and Multilingual Analysis](https://mlsanity.com/paper/2010.04543/) (2020)
- [Revisiting Machine Translation for Cross-lingual Classification](https://mlsanity.com/paper/2305.14240/) (2023)
- [Massively Multilingual Word Embeddings](https://mlsanity.com/paper/1602.01925/) (2016)
- [Toxicity in Multilingual Machine Translation at Scale](https://mlsanity.com/paper/2210.03070/) (2022)
- [Modelling Latent Translations for Cross-Lingual Transfer](https://mlsanity.com/paper/2107.11353/) (2021)
- [Multilingual and Explainable Text Detoxification with Parallel Corpora](https://mlsanity.com/paper/2412.11691/) (2024)
- [Toxicity Classification in Ukrainian](https://mlsanity.com/paper/2404.17841/) (2024)
- [Offensive Hebrew Corpus and Detection using BERT](https://mlsanity.com/paper/2309.02724/) (2023)
- [Multilingual and Multilabel Emotion Recognition using Virtual Adversarial Training](https://mlsanity.com/paper/2111.06181/) (2021)

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

- [Challenges and Recommendations for LLM-as-a-Judge in Multilingual Settings and for Low-Resource Languages](https://mlsanity.com/paper/2607.02235/) (2026)
- [Cross-Lingual Jailbreak Detection via Semantic Codebooks](https://mlsanity.com/paper/2604.25716/) (2026)
- [A Survey of Toxicity Detection and Mitigation Strategies for Multilingual Language Models](https://mlsanity.com/paper/2606.25380/) (2026)
- [LLM Safety Alignment in Low-Resource Languages: A Systematic Literature Review](https://mlsanity.com/paper/2608.14626/) (2026)

## Abstract (arXiv)

Multilingual toxicity detection remains a significant challenge due to the scarcity of training data and resources for many languages. While prior work has leveraged the translate-test paradigm to support cross-lingual transfer across a range of classification tasks, the utility of translation in supporting toxicity detection at scale remains unclear. In this work, we conduct a comprehensive comparison of translation-based and language-specific/multilingual classification pipelines. We find that translation-based pipelines consistently outperform out-of-distribution classifiers in 81.3% of cases (13 of 16 languages), with translation benefits strongly correlated with both the resource level of the target language and the quality of the machine translation (MT) system. Our analysis reveals that traditional classifiers outperform large language model (LLM) judges, with this advantage being particularly pronounced for low-resource languages, where translate-classify methods dominate translate-judge approaches in 6 out of 7 cases. We additionally show that MT-specific fine-tuning on LLMs yields lower refusal rates compared to standard instruction-tuned models, but it can negatively impact toxicity detection accuracy for low-resource languages. These findings offer actionable guidance for practitioners developing scalable multilingual content moderation systems.