Adaptive Detoxification: Safeguarding General Capabilities of LLMs through Toxicity-Aware Knowledge Editing

• URL: http://arxiv.org/abs/2505.22298v1
• Date: Wed, 28 May 2025 12:37:06 GMT
• Title: Adaptive Detoxification: Safeguarding General Capabilities of LLMs through Toxicity-Aware Knowledge Editing
• Authors: Yifan Lu, Jing Li, Yigeng Zhou, Yihui Zhang, Wenya Wang, Xiucheng Li, Meishan Zhang, Fangming Liu, Jun Yu, Min Zhang,
• Abstract summary: ToxEdit is a toxicity-aware knowledge editing approach.<n>It dynamically detects toxic activation patterns during forward propagation.<n>It then routes computations through adaptive inter-layer pathways to mitigate toxicity effectively.
• Score: 49.85884082568318
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) exhibit impressive language capabilities but remain vulnerable to malicious prompts and jailbreaking attacks. Existing knowledge editing methods for LLM detoxification face two major challenges. First, they often rely on entity-specific localization, making them ineffective against adversarial inputs without explicit entities. Second, these methods suffer from over-editing, where detoxified models reject legitimate queries, compromising overall performance. In this paper, we propose ToxEdit, a toxicity-aware knowledge editing approach that dynamically detects toxic activation patterns during forward propagation. It then routes computations through adaptive inter-layer pathways to mitigate toxicity effectively. This design ensures precise toxicity mitigation while preserving LLMs' general capabilities. To more accurately assess over-editing, we also enhance the SafeEdit benchmark by incorporating instruction-following evaluation tasks. Experimental results on multiple LLMs demonstrate that our ToxEdit outperforms previous state-of-the-art methods in both detoxification performance and safeguarding general capabilities of LLMs.

Related papers

• LLM Hypnosis: Exploiting User Feedback for Unauthorized Knowledge Injection to All Users [50.18141341939909]
  We describe a vulnerability in language models trained with user feedback.<n>A single user can persistently alter LM knowledge and behavior.<n>We show that this attack can be used to insert factual knowledge the model did not previously possess.
  arXiv  Detail & Related papers  (2025-07-03T17:55:40Z)
• Detoxification of Large Language Models through Output-layer Fusion with a Calibration Model [15.394714537797183]
  Existing approaches for Large language model (LLM) detoxification rely on training on large-scale non-toxic or human-annotated preference data.<n>We propose a compact, pre-trained calibration model that guides the detoxification process of a target LLM via a lightweight intervention in its generation pipeline.
  arXiv  Detail & Related papers  (2025-06-02T02:36:32Z)
• Breaking Bad Tokens: Detoxification of LLMs Using Sparse Autoencoders [3.6367827664262715]
  Large language models (LLMs) are now ubiquitous in user-facing applications, yet they still generate undesirable toxic outputs.<n>We leverage sparse autoencoders (SAEs) to identify toxicity-related directions in the residual stream of models and perform targeted activation steering.
  arXiv  Detail & Related papers  (2025-05-20T15:55:31Z)
• Large Language Models can be Strong Self-Detoxifiers [82.6594169242814]
  Self-disciplined Autoregressive Sampling (SASA) is a lightweight controlled decoding algorithm for toxicity reduction of large language models (LLMs) SASA tracks the margin of the current output to steer the generation away from the toxic subspace, by adjusting the autoregressive sampling strategy. evaluated on LLMs of different scale and nature, namely Llama-3.1-Instruct (8B), Llama-2 (7B), and GPT2-L models with the RealToxicityPrompts, BOLD, and AttaQ benchmarks.
  arXiv  Detail & Related papers  (2024-10-04T17:45:15Z)
• Model Surgery: Modulating LLM's Behavior Via Simple Parameter Editing [63.20133320524577]
  We show that editing a small subset of parameters can effectively modulate specific behaviors of large language models (LLMs)<n>Our approach achieves reductions of up to 90.0% in toxicity on the RealToxicityPrompts dataset and 49.2% on ToxiGen.
  arXiv  Detail & Related papers  (2024-07-11T17:52:03Z)
• Detoxifying Large Language Models via Knowledge Editing [57.0669577257301]
  This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs) We construct a benchmark, SafeEdit, which covers nine unsafe categories with various powerful attack prompts. We conduct experiments with several knowledge editing approaches, indicating that knowledge editing has the potential to detoxify LLMs with a limited impact on general performance efficiently.
  arXiv  Detail & Related papers  (2024-03-21T15:18:30Z)
• Unveiling the Implicit Toxicity in Large Language Models [77.90933074675543]
  The open-endedness of large language models (LLMs) combined with their impressive capabilities may lead to new safety issues when being exploited for malicious use. We show that LLMs can generate diverse implicit toxic outputs that are exceptionally difficult to detect via simply zero-shot prompting. We propose a reinforcement learning (RL) based attacking method to further induce the implicit toxicity in LLMs.
  arXiv  Detail & Related papers  (2023-11-29T06:42:36Z)
• Self-Detoxifying Language Models via Toxification Reversal [11.238212967733165]
  Language model detoxification aims to minimize the risk of generating offensive or harmful content in pretrained language models (PLMs) We propose a more lightweight approach that enables the PLM itself to achieve "self-detoxification" Our method is built upon the observation that prepending a negative steering prompt can effectively induce PLMs to generate toxic content.
  arXiv  Detail & Related papers  (2023-10-14T12:51:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.