GloSS over Toxicity: Understanding and Mitigating Toxicity in LLMs via Global Toxic Subspace

• URL: http://arxiv.org/abs/2505.17078v1
• Date: Tue, 20 May 2025 08:29:11 GMT
• Title: GloSS over Toxicity: Understanding and Mitigating Toxicity in LLMs via Global Toxic Subspace
• Authors: Zenghao Duan, Zhiyi Yin, Zhichao Shi, Liang Pang, Shaoling Jing, Jiayi Wu, Yu Yan, Huawei Shen, Xueqi Cheng,
• Abstract summary: This paper investigates the underlying mechanisms of toxicity generation in Large Language Models (LLMs)<n>We propose GloSS (Global Toxic Subspace Suppression), a lightweight, four-stage method that mitigates toxicity by identifying and removing the global toxic subspace from the parameters of FFN.
• Score: 62.68664365246247
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper investigates the underlying mechanisms of toxicity generation in Large Language Models (LLMs) and proposes an effective detoxification approach. Prior work typically considers the Feed-Forward Network (FFN) as the main source of toxicity, representing toxic regions as a set of toxic vectors or layer-wise subspaces. However, our in-depth analysis reveals that the global toxic subspace offers a more effective and comprehensive representation of toxic region within the model. Building on this insight, we propose GloSS (Global Toxic Subspace Suppression), a lightweight, four-stage method that mitigates toxicity by identifying and removing the global toxic subspace from the parameters of FFN. Experiments across a range of LLMs show that GloSS achieves state-of-the-art detoxification performance while preserving the models general capabilities, without requiring large-scale data or model retraining.

Related papers

• Aligned Probing: Relating Toxic Behavior and Model Internals [66.49887503194101]
  We introduce aligned probing, a novel interpretability framework that aligns the behavior of language models (LMs)<n>Using this framework, we examine over 20 OLMo, Llama, and Mistral models, bridging behavioral and internal perspectives for toxicity for the first time.<n>Our results show that LMs strongly encode information about the toxicity level of inputs and subsequent outputs, particularly in lower layers.
  arXiv  Detail & Related papers  (2025-03-17T17:23:50Z)
• 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)
• PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models [27.996123856250065]
  Existing toxicity benchmarks are overwhelmingly focused on English. We introduce PolygloToxicityPrompts (PTP), the first large-scale multilingual toxicity evaluation benchmark of 425K naturally occurring prompts spanning 17 languages.
  arXiv  Detail & Related papers  (2024-05-15T14:22:33Z)
• 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)
• Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metric [16.423707276483178]
  We introduce a robust metric grounded on Large Language Models (LLMs) to flexibly measure toxicity according to the given definition. Our results demonstrate outstanding performance in measuring toxicity within verified factors, improving on conventional metrics by 12 points in the F1 score.
  arXiv  Detail & Related papers  (2024-02-10T07:55:27Z)
• 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)
• Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmarks [18.44630180661091]
  Existing datasets lack fine-grained annotation of toxic types and expressions. It is crucial to introduce lexical knowledge to detect the toxicity of posts. In this paper, we facilitate the fine-grained detection of Chinese toxic language.
  arXiv  Detail & Related papers  (2023-05-08T03:50:38Z)
• Simple Text Detoxification by Identifying a Linear Toxic Subspace in Language Model Embeddings [8.720903734757627]
  Large pre-trained language models are often trained on large volumes of internet data, some of which may contain toxic or abusive language. Current methods aim to prevent toxic features from appearing generated text. We hypothesize the existence of a low-dimensional toxic subspace in the latent space of pre-trained language models.
  arXiv  Detail & Related papers  (2021-12-15T18:54:34Z)
• RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models [93.151822563361]
  Pretrained neural language models (LMs) are prone to generating racist, sexist, or otherwise toxic language which hinders their safe deployment. We investigate the extent to which pretrained LMs can be prompted to generate toxic language, and the effectiveness of controllable text generation algorithms at preventing such toxic degeneration.
  arXiv  Detail & Related papers  (2020-09-24T03:17:19Z)

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.