Griffin Floto, Mohammad Mahdi Abdollah Pour, Parsa Farinneya, Zhenwei Tang, Ali Pesaranghader, Manasa Bharadwaj, Scott Sanner
| Challenge: | Existing text generation models that reduce toxicity of toxic text are inadequate for text detoxification tasks. |
| Approach: | They propose a conditional and unconditional diffusion model for text detoxification . conditional model takes toxic text as condition and reduces its toxicity . experimental results show the model achieves human-level fluency . |
| Outcome: | The proposed model reduces toxic text and produces diverse sentences . it can be used to train other models and ensure fluency . |
Similar Papers
MultiParaDetox: Extending Text Detoxification with Parallel Data to New Languages (2024.naacl-short)
Copied to clipboard
| Challenge: | Text detoxification is a textual style transfer task where a toxic text is paraphrased to the neutral register. |
| Approach: | They propose to extend ParaDetox pipeline to multiple languages to automate parallel detoxification corpus collection. |
| Outcome: | The proposed methods have been used in toxic speech combating and toxic speech fighting tasks. |
CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) frequently generate toxic content, posing significant risks for safe deployment. |
| Approach: | They propose a framework that identifies and intervenes on the specific attention heads causally responsible for toxic generation. |
| Outcome: | The proposed framework reduces toxic generation by 5.34% while preserving linguistic fluency and speeding up head selection. |
Text Detoxification using Large Pre-trained Neural Models (2021.emnlp-main)
Copied to clipboard
David Dale, Anton Voronov, Daryna Dementieva, Varvara Logacheva, Olga Kozlova, Nikita Semenov, Alexander Panchenko
| Challenge: | Existing studies on text detoxification cast this task as style transfer . text detox requires better preservation of the original meaning, authors argue . |
| Approach: | They propose two unsupervised methods for eliminating toxicity in text . they use a paraphraser guided by style-trained language models to keep the text content . |
| Outcome: | The proposed methods yield new SOTA results. |
XDetox: Text Detoxification with Token-Level Toxicity Explanations (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for mitigating toxic content are black-box approaches, which results in limitations in modifying toxic tokens. |
| Approach: | They propose a method that integrates token-level toxicity explanations with the masking and infilling detoxification processes. |
| Outcome: | The proposed method outperforms baseline methods in fluency and toxicity reduction. |
Exploring Cross-lingual Text Detoxification with Large Multilingual Language Models. (2022.acl-srw)
Copied to clipboard
| Challenge: | Existing methods of textual style transfer are monolingual i.e. designed to work in one exact language. |
| Approach: | They propose to make large multilingual models capable of performing multilingual style transfer without direct fine-tuning in a given language. |
| Outcome: | The proposed model can generate text in a given language without fine-tuning and is able to perform cross-lingual detoxification without direct fine- tuning. |
DetoxLLM: A Framework for Detoxification with Explanations (2024.emnlp-main)
Copied to clipboard
| Challenge: | DetoxLLM is a comprehensive end-to-end detoxification framework for toxic language. |
| Approach: | They propose a comprehensive end-to-end detoxification framework that tackles toxic language across platforms. |
| Outcome: | The proposed detoxification framework outperforms the SoTA model on human-annotated parallel corpus and offers explanation to promote transparency and trustworthiness. |
Detoxifying Large Language Models via the Diversity of Toxic Samples (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for analyzing and utilizing toxic samples are limited . current methods fail to fully harness their potential . |
| Approach: | They propose a diverse detoxification framework that leverages toxic samples' diversity . they propose MPSG strategy and SC-DPO approach to elicit personalized toxic responses . |
| Outcome: | The proposed framework could be used to optimize large language models for user safety . it incorporates two components: MPSG strategy and SC-DPO approach . |
SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to multilingual text detoxification are hampered by the scarcity of parallel multilingual datasets. |
| Approach: | They propose a pipeline for the generation of multilingual parallel detoxification data and a dataset for SynthDetoxM which is manually generated and rewritten with open-source LLMs. |
| Outcome: | The proposed pipeline outperforms human-annotated datasets even in data limited setting. |
Self-Detoxifying Language Models via Toxification Reversal (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to generate toxic content in pretrained language models are resource-intensive and require additional components. |
| Approach: | They propose a method that enables the PLM itself to achieve "self-detoxification" they identify the toxification direction from the normal generation process to the one prompted with the negative prefix and then steer the generation to the reverse direction by manipulating the information movement within the attention layers. |
| Outcome: | The proposed method can achieve comparable performance with state-of-the-art methods without any fine-tuning or extra components. |
Contrastive Perplexity for Controlled Generation: An Application in Detoxifying Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to generate toxic content by large language models are based on pipelines . current approaches focus on preserving performance while effectively mitigating toxicity . |
| Approach: | They propose a framework for implicit knowledge editing and controlled text generation by using hard negatives. |
| Outcome: | The proposed framework significantly reduces toxic generation while maintaining strong performance on downstream tasks. |