个性化与安全的交汇：个性化大语言模型中的机制、风险与缓解措施
Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
AI总结：
本文首次对个性化大语言模型进行安全导向的综述，从用户表征、个性化范式和评估三个维度组织，提出统一的安全风险分类，并分析各范式下的脆弱性及缓解策略。
摘要：
大语言模型通过适应用户偏好、上下文和长期历史记录，实现了日益个性化的交互。然而，实现个性化的机制也以现有文献未系统处理的方式扩展了安全领域。现有综述通常只关注个性化或安全，而忽略了它们的交叉。我们提出了首个全面的、安全导向的个性化大语言模型综述。我们沿三个维度组织个性化——用户表征、个性化范式和评估——并引入统一的安全风险分类。在表征层面，我们分析了不同用户表征带来的风险。在主流个性化范式中，我们描述了提示、检索增强、参数微调、强化学习、混合专家、剪枝、智能体框架和多模态个性化中固有的脆弱性，并综合了模型生命周期中的缓解策略。除了这些细粒度风险，我们还描述了由个性化适应产生的范式无关的安全风险。我们进一步总结了个性化数据集和评估方法。通过OpenClaw的案例研究，我们分析了个性化智能体生态系统中的部署趋势。我们的分析揭示了现有研究中的三个结构性不足：安全被评估为与用户无关而非关系性的，个性化技术被孤立分析而非组合分析，评估框架无法捕捉新兴的长期风险。通过联合检查个性化表征、个性化范式、安全风险、防御和评估方法，我们为开发安全的个性化大语言模型提供了一个统一框架，并强调了未来研究的关键方向。
英文摘要：
Large Language Models (LLMs) have enabled increasingly personalized interactions by adapting to users' preferences, contexts, and long-term histories. However, the mechanisms that enable personalization also expand the safety landscape in ways not systematically addressed by existing literature. Existing reviews typically focus either on personalization or safety, leaving their intersection largely unexplored. We present the first comprehensive, safety-aware review of personalized LLMs. We organize personalization along three dimensions-user representation, personalization paradigm, and evaluation-and introduce a unified taxonomy of safety risks. At the representation level, we analyze risks arising from diverse user representations. Across mainstream personalization paradigms, we delineate vulnerabilities inherent to prompting, retrieval augmentation, parameter fine-tuning, reinforcement learning, Mixture-of-Experts (MoE), pruning, agent frameworks, and multimodal personalization, and synthesize mitigation strategies across the model lifecycle. Beyond these fine-grained risks, we characterize paradigm-agnostic safety risks arising from personalized adaptation. We further summarize personalized datasets and evaluation methodologies. Through a case study of OpenClaw, we analyze deployment trends in personalized agent ecosystems. Our analysis reveals three structural inadequacies in existing research: safety is evaluated as user-invariant rather than relational, personalization techniques are analyzed in isolation rather than in composition, and evaluation frameworks cannot capture emergent long-term risks. By jointly examining personalized representations, personalization paradigms, safety risks, defenses, and evaluation methods, we provide a unified framework for developing safe personalized LLMs and highlight key directions for future research.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
