{"slug": "beyond-a-global-norm-personalizing-toxicity-sensitivity-in-language-models", "title": "Beyond a Global Norm: Personalizing Toxicity Sensitivity in Language Models Without Retraining", "summary": "A study from arXiv (2607.23175v1) presents the first comparative evaluation of training-free methods for aligning language models to user-specific toxicity sensitivities, finding that all methods reduce alignment error by 28-47% but reveal a fundamental trade-off between alignment effectiveness, personalization, and language quality.", "body_md": "arXiv:2607.23175v1 Announce Type: new\nAbstract: Reducing toxicity is often framed as a global alignment problem, yet perceptions of harmful language are subjective and context-dependent. We present the first comparative evaluation of training-free methods for aligning language generation to user-specific toxicity sensitivities across three inference-time intervention stages: pre-decoding (prompt conditioning and rewriting), in-decoding (token, logit, and representation steering), and post-decoding (candidate re-ranking). Evaluated against toxicity sensitivity targets derived from the PRISM dataset, all methods reduce alignment error by 28-47%. However, the results reveal a fundamental trade-off between alignment effectiveness, personalization, and general language quality, showing how toxicity sensitivity alignment is an inherently multi-objective problem.", "url": "https://wpnews.pro/news/beyond-a-global-norm-personalizing-toxicity-sensitivity-in-language-models", "canonical_source": "https://arxiv.org/abs/2607.23175", "published_at": "2026-07-28 04:00:00+00:00", "updated_at": "2026-07-28 04:26:17.023328+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-safety", "ai-ethics", "natural-language-processing"], "entities": ["arXiv", "PRISM dataset"], "alternates": {"html": "https://wpnews.pro/news/beyond-a-global-norm-personalizing-toxicity-sensitivity-in-language-models", "markdown": "https://wpnews.pro/news/beyond-a-global-norm-personalizing-toxicity-sensitivity-in-language-models.md", "text": "https://wpnews.pro/news/beyond-a-global-norm-personalizing-toxicity-sensitivity-in-language-models.txt", "jsonld": "https://wpnews.pro/news/beyond-a-global-norm-personalizing-toxicity-sensitivity-in-language-models.jsonld"}}