# Chinese Toxic Language Mitigation via Sentiment Polarity Consistent Rewrites

arXiv: https://arxiv.org/abs/2505.15297 · HTML: https://mlsanity.com/paper/2505.15297/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 21 May 2025.

## Builds on (most cited first, 12 of 12)

- [COLD: A Benchmark for Chinese Offensive Language Detection](https://mlsanity.com/paper/2201.06025/) (2022)
- [Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmarks](https://mlsanity.com/paper/2305.04446/) (2023)
- [SWSR: A Chinese Dataset and Lexicon for Online Sexism Detection](https://mlsanity.com/paper/2108.03070/) (2021)
- [Towards Identifying Social Bias in Dialog Systems: Frame, Datasets, and Benchmarks](https://mlsanity.com/paper/2202.08011/) (2022)
- [ROSE Doesn't Do That: Boosting the Safety of Instruction-Tuned Large Language Models with Reverse Prompt Contrastive Decoding](https://mlsanity.com/paper/2402.11889/) (2024)
- [ToxiCloakCN: Evaluating Robustness of Offensive Language Detection in Chinese with Cloaking Perturbations](https://mlsanity.com/paper/2406.12223/) (2024)
- [Walking in Others' Shoes: How Perspective-Taking Guides Large Language Models in Reducing Toxicity and Bias](https://mlsanity.com/paper/2407.15366/) (2024)
- [Take Care of Your Prompt Bias! Investigating and Mitigating Prompt Bias in Factual Knowledge Extraction](https://mlsanity.com/paper/2403.09963/) (2024)
- [Multilingual and Explainable Text Detoxification with Parallel Corpora](https://mlsanity.com/paper/2412.11691/) (2024)
- [Tox-BART: Leveraging Toxicity Attributes for Explanation Generation of Implicit Hate Speech](https://mlsanity.com/paper/2406.03953/) (2024)
- [Toxicity Detection is NOT all you Need: Measuring the Gaps to Supporting Volunteer Content Moderators](https://mlsanity.com/paper/2311.07879/) (2023)
- [SCCD: A Session-based Dataset for Chinese Cyberbullying Detection](https://mlsanity.com/paper/2501.15042/) (2025)

## Built on by (most cited first, 2 of 2)

- [Content Fuzzing for Escaping Information Cocoons on Digital Social Media](https://mlsanity.com/paper/2604.05461/) (2026)
- [Detoxify: A framework for abusive text transformation using LLMs](https://mlsanity.com/paper/2507.10177/) (2025)

## Abstract (arXiv)

Detoxifying offensive language while preserving the speaker's original intent is a challenging yet critical goal for improving the quality of online interactions. Although large language models (LLMs) show promise in rewriting toxic content, they often default to overly polite rewrites, distorting the emotional tone and communicative intent. This problem is especially acute in Chinese, where toxicity often arises implicitly through emojis, homophones, or discourse context. We present ToxiRewriteCN, the first Chinese detoxification dataset explicitly designed to preserve sentiment polarity. The dataset comprises 1,556 carefully annotated triplets, each containing a toxic sentence, a sentiment-aligned non-toxic rewrite, and labeled toxic spans. It covers five real-world scenarios: standard expressions, emoji-induced and homophonic toxicity, as well as single-turn and multi-turn dialogues. We evaluate 17 LLMs, including commercial and open-source models with variant architectures, across four dimensions: detoxification accuracy, fluency, content preservation, and sentiment polarity. Results show that while commercial and MoE models perform best overall, all models struggle to balance safety with emotional fidelity in more subtle or context-heavy settings such as emoji, homophone, and dialogue-based inputs. We release ToxiRewriteCN to support future research on controllable, sentiment-aware detoxification for Chinese.