CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification

• URL: http://arxiv.org/abs/2604.14602v1
• Date: Thu, 16 Apr 2026 04:19:48 GMT
• Title: CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification
• Abstract summary: Large language models (LLMs) frequently generate toxic content, posing significant risks for safe deployment.<n>We propose CAUSALDETOX, a framework that identifies and intervenes on the specific attention heads causally responsible for toxic generation.
• Score: 10.438004034144884
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) frequently generate toxic content, posing significant risks for safe deployment. Current mitigation strategies often degrade generation quality or require costly human annotation. We propose CAUSALDETOX, a framework that identifies and intervenes on the specific attention heads causally responsible for toxic generation. Using the Probability of Necessity and Sufficiency (PNS), we isolate a minimal set of heads that are necessary and sufficient for toxicity. We utilize these components via two complementary strategies: (1) Local Inference-Time Intervention, which constructs dynamic, input-specific steering vectors for context-aware detoxification, and (2) PNS-Guided Fine-Tuning, which permanently unlearns toxic representations. We also introduce PARATOX, a novel benchmark of aligned toxic/non-toxic sentence pairs enabling controlled counterfactual evaluation. Experiments on ToxiGen, ImplicitHate, and ParaDetox show that CAUSALDETOX achieves up to 5.34% greater toxicity reduction compared to baselines while preserving linguistic fluency, and offers a 7x speedup in head selection.

Related papers

• Where Does Toxicity Live? Mechanistic Localization and Targeted Suppression in Language Models [1.2040549525117266]
  Large language models frequently generate toxic, hateful, or harmful content.<n>We introduce Meow2X and TRNE, two complementary retraining-free frameworks that localize toxicity to specific layers and neurons.<n>Our analysis reveals that toxicity is disproportionately encoded in early gradient layers, varies across architectures, and is systematically underestimated by single-evaluator setups.
  arXiv  Detail & Related papers  (2026-05-27T05:41:19Z)
• Measuring and Mitigating Toxicity in Large Language Models: A Comprehensive Replication Study [0.0]
  Large Language Models (LLMs) trained on web-scale corpora inherently absorb toxic patterns from their training data.<n>Decoding-time Experts (DExperts) is an inference-time mitigation technique that steers generation without requiring model retraining.<n>This study contributes to the growing body of work on AI safety by highlighting the gap between explicit and implicit toxicity mitigation.
  arXiv  Detail & Related papers  (2026-05-13T20:12:30Z)
• Detoxification for LLM: From Dataset Itself [69.09071947137926]
  Existing detoxification methods for large language models mainly focus on post-training stage or inference time, while few tackle the source of toxicity, namely, the dataset itself.<n>We attempt to detoxify directly on raw corpora with SoCD (Soft Contrastive Decoding), which guides an LLM to localize and rewrite toxic spans in raw data while preserving semantics.<n>On GPT2-XL, HSPD attains state-of-the-art detoxification, reducing Toxicity Probability (TP) from 0.42 to 0.18 and Maximum Expected Toxicity (EMT) from 0.43 to 0.20.
  arXiv  Detail & Related papers  (2026-04-21T06:12:29Z)
• Do Prompts Guarantee Safety? Mitigating Toxicity from LLM Generations through Subspace Intervention [6.808534332444413]
  Large Language Models (LLMs) are powerful text generators.<n>LLMs can produce toxic or harmful content even when given seemingly harmless prompts.<n>This presents a serious safety challenge and can cause real-world harm.
  arXiv  Detail & Related papers  (2026-02-06T11:33:17Z)
• Projecting Out the Malice: A Global Subspace Approach to LLM Detoxification [73.77171973106567]
  Large language models (LLMs) exhibit exceptional performance but pose inherent risks of generating toxic content.<n>Traditional methods fail to eliminate underlying toxic regions in parameters, leaving models vulnerable to adversarial attacks.<n>We propose GLOSS, a lightweight method that mitigates toxicity by identifying and eliminating this global subspace from FFN parameters.
  arXiv  Detail & Related papers  (2026-01-09T09:34:53Z)
• Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing [77.75609817898035]
  Large Language Models (LLMs) have demonstrated impressive performance across various tasks, yet they remain vulnerable to generating toxic content.<n>We propose textscAutoregressive textscReward textscGuided textscRe presentation textscEditing (ARGRE)<n>ARGRE explicitly models toxicity transitions within the latent representation space, enabling stable and precise reward-guided editing.
  arXiv  Detail & Related papers  (2025-09-24T03:40:32Z)
• Redefining Experts: Interpretable Decomposition of Language Models for Toxicity Mitigation [12.58703387927632]
  We investigate the stability of neuron-level toxicity indicators, the advantages of structural (layer-wise) representations, and the interpretability of mechanisms driving toxic generation.<n>We propose a novel principled intervention technique, EigenShift, based on eigen-decomposition of the language model's final output layer.
  arXiv  Detail & Related papers  (2025-09-20T12:21:52Z)
• GloSS over Toxicity: Understanding and Mitigating Toxicity in LLMs via Global Toxic Subspace [62.68664365246247]
  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.
  arXiv  Detail & Related papers  (2025-05-20T08:29:11Z)
• Mitigating Text Toxicity with Counterfactual Generation [0.3250512744763586]
  Toxicity mitigation consists in rephrasing text in order to remove harmful meaning.<n>Current methods fail to detoxify text while preserving the initial non-toxic meaning.<n>This work is the first to bridge the gap between counterfactual generation and text detoxification.
  arXiv  Detail & Related papers  (2024-05-16T09:52:21Z)
• Parameter-Efficient Detoxification with Contrastive Decoding [78.5124331048714]
  We introduce Detoxification Generator (DETOXIGEN), an inference-time algorithm that steers the generation away from unwanted styles. During the actual generation, we use the trained detoxifier to produce undesirable tokens for the generator to contrast against at each decoding step. We find that it significantly outperforms previous approaches in detoxification metrics while not compromising on the generation quality.
  arXiv  Detail & Related papers  (2024-01-13T01:46:20Z)
• 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)
• 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.