# Detoxify: A framework for abusive text transformation using LLMs

arXiv: https://arxiv.org/abs/2507.10177 · HTML: https://mlsanity.com/paper/2507.10177/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 31 of its 79 references that are in the map.
- First posted on arXiv 14 Jul 2025.

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

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [MPNet: Masked and Permuted Pre-training for Language Understanding](https://mlsanity.com/paper/2004.09297/) (2020)
- [A Comprehensive Overview of Large Language Models](https://mlsanity.com/paper/2307.06435/) (2023)
- [An Introduction to Convolutional Neural Networks](https://mlsanity.com/paper/1511.08458/) (2015)
- [Med-BERT: pre-trained contextualized embeddings on large-scale structured electronic health records for disease prediction](https://mlsanity.com/paper/2005.12833/) (2020)
- [Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior](https://mlsanity.com/paper/1802.00393/) (2018)
- [Multimodal Deep Learning](https://mlsanity.com/paper/2301.04856/) (2023)
- [Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection](https://mlsanity.com/paper/2012.15761/) (2020)
- [HateBERT: Retraining BERT for Abusive Language Detection in English](https://mlsanity.com/paper/2010.12472/) (2020)
- [SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media (OffensEval 2020)](https://mlsanity.com/paper/2006.07235/) (2020)
- [ChatGPT is not all you need. A State of the Art Review of large Generative AI models](https://mlsanity.com/paper/2301.04655/) (2023)
- [Text Detoxification using Large Pre-trained Neural Models](https://mlsanity.com/paper/2109.08914/) (2021)
- [Detoxifying Text with MaRCo: Controllable Revision with Experts and Anti-Experts](https://mlsanity.com/paper/2212.10543/) (2022)
- [Tackling Online Abuse: A Survey of Automated Abuse Detection Methods](https://mlsanity.com/paper/1908.06024/) (2019)
- [Robust Hate Speech Detection in Social Media: A Cross-Dataset Empirical Evaluation](https://mlsanity.com/paper/2307.01680/) (2023)
- [A systematic review of Hate Speech automatic detection using Natural Language Processing](https://mlsanity.com/paper/2106.00742/) (2021)
- [Detection of Hate Speech using BERT and Hate Speech Word Embedding with Deep Model](https://mlsanity.com/paper/2111.01515/) (2021)
- [COVID-19 sentiment analysis via deep learning during the rise of novel cases](https://mlsanity.com/paper/2104.10662/) (2021)
- [Re-balancing Variational Autoencoder Loss for Molecule Sequence Generation](https://mlsanity.com/paper/1910.00698/) (2019)
- [Semantic and sentiment analysis of selected Bhagavad Gita translations using BERT-based language framework](https://mlsanity.com/paper/2201.03115/) (2022)
- [DetoxLLM: A Framework for Detoxification with Explanations](https://mlsanity.com/paper/2402.15951/) (2024)
- [Don't Take This Out of Context! On the Need for Contextual Models and Evaluations for Stylistic Rewriting](https://mlsanity.com/paper/2305.14755/) (2023)
- [Estimating Social Influence from Observational Data](https://mlsanity.com/paper/2204.01633/) (2022)
- [Detecting LLM-assisted writing in scientific communication: Are we there yet?](https://mlsanity.com/paper/2401.16807/) (2024)
- [An evaluation of LLMs and Google Translate for translation of selected Indian languages via sentiment and semantic analyses](https://mlsanity.com/paper/2503.21393/) (2025)

## Abstract (arXiv)

Although Large Language Models (LLMs) have demonstrated significant advancements in natural language processing tasks, their effectiveness in the classification and transformation of abusive text into non-abusive versions remains an area for exploration. In this study, we present Detoxify: a framework that employs LLMs to transform abusive text (tweets and reviews) containing hate speech and profanity into non-abusive text while retaining the original intent. We evaluate the performance of four state-of-the-art LLMs, such as Gemini, GPT-4o, DeekSeek and Groq, on their ability to identify abusive text. We aim to transform and obtain a text that is clean of abusive and inappropriate content, but maintains a similar level of sentiment and semantics, i.e. the transformed text needs to maintain its message. Afterwards, we evaluate the raw and transformed datasets with sentiment analysis and semantic analysis. Our results show Groq provides vastly different results when compared with other LLMs. We have identified similarities between GPT-4o and DeepSeek. Groq stood out as the most distinct, as it often restructured sentences with excessive positive phrasing, with the original context lost or altered.