# Fine-Grained Detoxification via Instance-Level Prefixes for Large Language Models

arXiv: https://arxiv.org/abs/2402.15202 · HTML: https://mlsanity.com/paper/2402.15202/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 47 references that are in the map.
- First posted on arXiv 23 Feb 2024.

## Builds on (most cited first, 30 of 35)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [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)
- [Scaling Language Models: Methods, Analysis & Insights from Training Gopher](https://mlsanity.com/paper/2112.11446/) (2021)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://mlsanity.com/paper/1909.05858/) (2019)
- [Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution](https://mlsanity.com/paper/2202.10054/) (2022)
- [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://mlsanity.com/paper/2311.03099/) (2023)
- [Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study](https://mlsanity.com/paper/2305.13860/) (2023)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [GeDi: Generative Discriminator Guided Sequence Generation](https://mlsanity.com/paper/2009.06367/) (2020)
- [Toxicity in ChatGPT: Analyzing Persona-assigned Language Models](https://mlsanity.com/paper/2304.05335/) (2023)
- [Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP](https://mlsanity.com/paper/2103.00453/) (2021)
- [In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering](https://mlsanity.com/paper/2311.06668/) (2023)
- [Composing Parameter-Efficient Modules with Arithmetic Operations](https://mlsanity.com/paper/2306.14870/) (2023)
- [On Prompt-Driven Safeguarding for Large Language Models](https://mlsanity.com/paper/2401.18018/) (2024)
- [Lift Yourself Up: Retrieval-augmented Text Generation with Self Memory](https://mlsanity.com/paper/2305.02437/) (2023)
- [Controlled Text Generation with Natural Language Instructions](https://mlsanity.com/paper/2304.14293/) (2023)
- [Model Merging by Uncertainty-Based Gradient Matching](https://mlsanity.com/paper/2310.12808/) (2023)
- [Controllable Natural Language Generation with Contrastive Prefixes](https://mlsanity.com/paper/2202.13257/) (2022)
- [On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmark](https://mlsanity.com/paper/2110.08466/) (2021)
- [Exploring the Limits of Domain-Adaptive Training for Detoxifying Large-Scale Language Models](https://mlsanity.com/paper/2202.04173/) (2022)
- [Self-Detoxifying Language Models via Toxification Reversal](https://mlsanity.com/paper/2310.09573/) (2023)
- [Detoxifying Text with MaRCo: Controllable Revision with Experts and Anti-Experts](https://mlsanity.com/paper/2212.10543/) (2022)
- [Separate the Wheat from the Chaff: Model Deficiency Unlearning via Parameter-Efficient Module Operation](https://mlsanity.com/paper/2308.08090/) (2023)
- [FFT: Towards Harmlessness Evaluation and Analysis for LLMs with Factuality, Fairness, Toxicity](https://mlsanity.com/paper/2311.18580/) (2023)

## Abstract (arXiv)

Impressive results have been achieved in natural language processing (NLP) tasks through the training of large language models (LLMs). However, these models occasionally produce toxic content such as insults, threats, and profanity in response to certain prompts, thereby constraining their practical utility. To tackle this issue, various finetuning-based and decoding-based approaches have been utilized to mitigate toxicity. However, these methods typically necessitate additional costs such as high-quality training data or auxiliary models. In this paper, we propose fine-grained detoxification via instance-level prefixes (FGDILP) to mitigate toxic text without additional cost. Specifically, FGDILP contrasts the contextualized representation in attention space using a positive prefix-prepended prompt against multiple negative prefix-prepended prompts at the instance level. This allows for constructing fine-grained subtoxicity vectors, which enables collaborative detoxification by fusing them to correct the normal generation process when provided with a raw prompt. We validate that FGDILP enables controlled text generation with regard to toxicity at both the utterance and context levels. Our method surpasses prompt-based baselines in detoxification, although at a slight cost to generation fluency and diversity.