2606.09038v1 Jun 08, 2026 cs.AI

Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs

Junlan Feng
Junlan Feng
Citations: 5
h-index: 1
Yanyan Luo
Yanyan Luo
Citations: 21
h-index: 2
Xue Han
Xue Han
Citations: 134
h-index: 4
Ruiqiao Bai
Ruiqiao Bai
Citations: 135
h-index: 5
Chunxu Zhao
Chunxu Zhao
Citations: 20
h-index: 2
Xinyi Huang
Xinyi Huang
Citations: 73
h-index: 4
Yitong Wang
Yitong Wang
Citations: 65
h-index: 3
Qian Hu
Qian Hu
Citations: 12
h-index: 2
Qing Wang
Qing Wang
Citations: 32
h-index: 3
Jie Liu
Jie Liu
Citations: 5
h-index: 2
Cong Geng
Cong Geng
Citations: 206
h-index: 4
Lehao Xing
Lehao Xing
Citations: 132
h-index: 3
Peng Hu
Peng Hu
Citations: 66
h-index: 3

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.

0 Citations
0 Influential
2.5 Altmetric
12.5 Score
Original PDF

No Analysis Report Yet

This paper hasn't been analyzed by Gemini yet.

Log in to request an AI analysis.

댓글

댓글을 작성하려면 로그인하세요.

아직 댓글이 없습니다. 첫 번째 댓글을 남겨보세요!