# RTP-LX: Can LLMs Evaluate Toxicity in Multilingual Scenarios?

arXiv: https://arxiv.org/abs/2404.14397 · HTML: https://mlsanity.com/paper/2404.14397/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 22 Apr 2024.

## Builds on (most cited first, 24 of 24)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [The Woman Worked as a Babysitter: On Biases in Language Generation](https://mlsanity.com/paper/1909.01326/) (2019)
- [DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models](https://mlsanity.com/paper/2306.11698/) (2023)
- [BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation](https://mlsanity.com/paper/2101.11718/) (2021)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [ChatGPT Beyond English: Towards a Comprehensive Evaluation of Large Language Models in Multilingual Learning](https://mlsanity.com/paper/2304.05613/) (2023)
- [Challenges in Detoxifying Language Models](https://mlsanity.com/paper/2109.07445/) (2021)
- [Are Large Language Model-based Evaluators the Solution to Scaling Up Multilingual Evaluation?](https://mlsanity.com/paper/2309.07462/) (2023)
- [Hate Speech Detection and Racial Bias Mitigation in Social Media based on BERT model](https://mlsanity.com/paper/2008.06460/) (2020)
- [MEGAVERSE: Benchmarking Large Language Models Across Languages, Modalities, Models and Tasks](https://mlsanity.com/paper/2311.07463/) (2023)
- [Offensive Language and Hate Speech Detection for Danish](https://mlsanity.com/paper/1908.04531/) (2019)
- [Characteristics of Harmful Text: Towards Rigorous Benchmarking of Language Models](https://mlsanity.com/paper/2206.08325/) (2022)
- [Multi-task Learning for Multilingual Neural Machine Translation](https://mlsanity.com/paper/2010.02523/) (2020)
- [Toxic Language Detection in Social Media for Brazilian Portuguese: New Dataset and Multilingual Analysis](https://mlsanity.com/paper/2010.04543/) (2020)
- [BEEP! Korean Corpus of Online News Comments for Toxic Speech Detection](https://mlsanity.com/paper/2005.12503/) (2020)
- [Evaluating the Elementary Multilingual Capabilities of Large Language Models with MultiQ](https://mlsanity.com/paper/2403.03814/) (2024)
- [An Evaluation on Large Language Model Outputs: Discourse and Memorization](https://mlsanity.com/paper/2304.08637/) (2023)
- [Rethinking Generative Large Language Model Evaluation for Semantic Comprehension](https://mlsanity.com/paper/2403.07872/) (2024)
- [Comprehensive Evaluation of ChatGPT Reliability Through Multilingual Inquiries](https://mlsanity.com/paper/2312.10524/) (2023)
- [Offensive Hebrew Corpus and Detection using BERT](https://mlsanity.com/paper/2309.02724/) (2023)

## Built on by (most cited first, 30 of 36)

- [Qwen3Guard Technical Report](https://mlsanity.com/paper/2510.14276/) (2025)
- [A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment](https://mlsanity.com/paper/2504.15585/) (2025)
- [Survey of Cultural Awareness in Language Models: Text and Beyond](https://mlsanity.com/paper/2411.00860/) (2024)
- [Large Language Model Safety: A Holistic Survey](https://mlsanity.com/paper/2412.17686/) (2024)
- [PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages](https://mlsanity.com/paper/2504.04377/) (2025)
- [PARIKSHA: A Large-Scale Investigation of Human-LLM Evaluator Agreement on Multilingual and Multi-Cultural Data](https://mlsanity.com/paper/2406.15053/) (2024)
- [Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle](https://mlsanity.com/paper/2407.13833/) (2024)
- [Preference Tuning For Toxicity Mitigation Generalizes Across Languages](https://mlsanity.com/paper/2406.16235/) (2024)
- [CultureGuard: Towards Culturally-Aware Dataset and Guard Model for Multilingual Safety Applications](https://mlsanity.com/paper/2508.01710/) (2025)
- [LLM Can be a Dangerous Persuader: Empirical Study of Persuasion Safety in Large Language Models](https://mlsanity.com/paper/2504.10430/) (2025)
- [PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models](https://mlsanity.com/paper/2405.09373/) (2024)
- [Enhancing LLM Safety Through a Theoretical Minimax Game Lens](https://mlsanity.com/paper/2502.05163/) (2025)
- [Multilingual Large Language Models: A Systematic Survey](https://mlsanity.com/paper/2411.11072/) (2024)
- [Arabic Dataset for LLM Safeguard Evaluation](https://mlsanity.com/paper/2410.17040/) (2024)
- [Benchmarking LLM Guardrails in Handling Multilingual Toxicity](https://mlsanity.com/paper/2410.22153/) (2024)
- [MrGuard: A Multilingual Reasoning Guardrail for Universal LLM Safety](https://mlsanity.com/paper/2504.15241/) (2025)
- [DeMod: A Holistic Tool with Explainable Detection and Personalized Modification for Toxicity Censorship](https://mlsanity.com/paper/2411.01844/) (2024)
- [LLMs Lost in Translation: M-ALERT uncovers Cross-Linguistic Safety Inconsistencies](https://mlsanity.com/paper/2412.15035/) (2024)
- [Awes, Laws, and Flaws From Today's LLM Research](https://mlsanity.com/paper/2408.15409/) (2024)
- [Cross-Lingual Transfer of Debiasing and Detoxification in Multilingual LLMs: An Extensive Investigation](https://mlsanity.com/paper/2412.14050/) (2024)
- [Aligned Probing: Relating Toxic Behavior and Model Internals](https://mlsanity.com/paper/2503.13390/) (2025)
- [ToxicTone: A Mandarin Audio Dataset Annotated for Toxicity and Toxic Utterance Tonality](https://mlsanity.com/paper/2505.15773/) (2025)
- [The Thin Line Between Comprehension and Persuasion in LLMs](https://mlsanity.com/paper/2507.01936/) (2025)
- [Evaluating the Robustness of Large Language Model Safety Guardrails Against Adversarial Attacks](https://mlsanity.com/paper/2511.22047/) (2025)
- [Massively Multilingual Adaptation of Large Language Models Using Bilingual Translation Data](https://mlsanity.com/paper/2506.00469/) (2025)
- [Breaking mBad! Supervised Fine-tuning for Cross-Lingual Detoxification](https://mlsanity.com/paper/2505.16722/) (2025)
- [ToxiFrench: Benchmarking and Enhancing Language Models via CoT Fine-Tuning for French Toxicity Detection](https://mlsanity.com/paper/2508.11281/) (2025)
- [LASA: Language-Agnostic Semantic Alignment at the Semantic Bottleneck for LLM Safety](https://mlsanity.com/paper/2604.12710/) (2026)
- [Qorgau: Evaluating LLM Safety in Kazakh-Russian Bilingual Contexts](https://mlsanity.com/paper/2502.13640/) (2025)
- [Lost in Localization: Building RabakBench with Human-in-the-Loop Validation to Measure Multilingual Safety Gaps](https://mlsanity.com/paper/2507.05980/) (2025)

## Abstract (arXiv)

Large language models (LLMs) and small language models (SLMs) are being adopted at remarkable speed, although their safety still remains a serious concern. With the advent of multilingual S/LLMs, the question now becomes a matter of scale: can we expand multilingual safety evaluations of these models with the same velocity at which they are deployed? To this end, we introduce RTP-LX, a human-transcreated and human-annotated corpus of toxic prompts and outputs in 28 languages. RTP-LX follows participatory design practices, and a portion of the corpus is especially designed to detect culturally-specific toxic language. We evaluate 10 S/LLMs on their ability to detect toxic content in a culturally-sensitive, multilingual scenario. We find that, although they typically score acceptably in terms of accuracy, they have low agreement with human judges when scoring holistically the toxicity of a prompt; and have difficulty discerning harm in context-dependent scenarios, particularly with subtle-yet-harmful content (e.g. microaggressions, bias). We release this dataset to contribute to further reduce harmful uses of these models and improve their safe deployment.