mlsanityData to 7 Oct 2026

Is Safety Standard Same for Everyone? User-Specific Safety Evaluation of Large Language Models

arXiv 2502.15086 ↗First posted 20 Feb 2025cs.CLcited by 10

On the map: LLM safety and security → Safety alignment, refusal and guardrails

Abstract (arXiv)

As the use of large language model (LLM) agents continues to grow, their safety vulnerabilities have become increasingly evident. Extensive benchmarks evaluate various aspects of LLM safety by defining the safety relying heavily on general standards, overlooking user-specific standards. However, safety standards for LLM may vary based on a user-specific profiles rather than being universally consistent across all users. This raises a critical research question: Do LLM agents act safely when considering user-specific safety standards? Despite its importance for safe LLM use, no benchmark datasets currently exist to evaluate the user-specific safety of LLMs. To address this gap, we introduce U-SafeBench, a benchmark designed to assess user-specific aspect of LLM safety. Our evaluation of 20 widely used LLMs reveals current LLMs fail to act safely when considering user-specific safety standards, marking a new discovery in this field. To address this vulnerability, we propose a simple remedy based on chain-of-thought, demonstrating its effectiveness in improving user-specific safety. Our benchmark and code are available at https://github.com/yeonjun-in/U-SafeBench.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security. Builds on 26 of its 44 references that are in the map. First posted on arXiv 20 Feb 2025.

Builds on

26 of 44 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements11 Oct
  2. 2024ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming6 Apr

Other references · 24

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  4. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,175
  5. 2023A Survey on Large Language Model based Autonomous Agentscited by 2,203
  6. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,928
  7. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,880
  8. 2023Jailbroken: How Does LLM Safety Training Fail?cited by 1,565
  9. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agentscited by 1,197
  10. 2023Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversationscited by 1,124
  11. 2023Fine-Grained Human Feedback Gives Better Rewards for Language Model Trainingcited by 399
  12. 2024Agent-SafetyBench: Evaluating the Safety of LLM Agentscited by 280
12 more
  1. 2023Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Mergingcited by 271
  2. 2024SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusalcited by 224
  3. 2023Beyond Memorization: Violating Privacy Via Inference with Large Language Modelscited by 167
  4. 2024Personalized Language Modeling from Personalized Human Feedbackcited by 124
  5. 2024SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agentscited by 102
  6. 2023Red teaming ChatGPT via Jailbreaking: Bias, Robustness, Reliability and Toxicitycited by 86
  7. 2024Aligning to Thousands of Preferences via System Message Generalizationcited by 84
  8. 2023Everyone Deserves A Reward: Learning Customized Human Preferencescited by 42
  9. 2024Exploring Safety-Utility Trade-Offs in Personalized Language Modelscited by 26
  10. 2022Revision Transformers: Instructing Language Models to Change their Valuescited by 8
  11. 2022ILLUME: Rationalizing Vision-Language Models through Human Interactionscited by 5
  12. 2024Unveiling Performance Challenges of Large Language Models in Low-Resource Healthcare: A Demographic Fairness Perspectivecited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 5; 2026: 5.

10 papers.

  1. 2025Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agentscited by 20
  2. 2025PSG-Agent: Personality-Aware Safety Guardrail for LLM-based Agentscited by 12
  3. 2025R1-ACT: Efficient Reasoning Model Safety Alignment by Activating Safety Knowledgecited by 6
  4. 2026PluriHarms: Benchmarking the Full Spectrum of Human Judgments on AI Harmcited by 5
  5. 2025Reasoning over Boundaries: Enhancing Specification Alignment via Test-time Deliberationcited by 3
  6. 2026Safety is Contextual, LLM-Judges Are Not: Navigating the Rigid Priors of Evaluatorscited by 2
  7. 2025Beyond Context: Large Language Models' Failure to Grasp Users' Intentcited by 1
  8. 2026When Seeing Overrides Knowing: Visual Dominance and Deferral-Based Method for Personalized Safety in VLMscited by 0
  9. 2026MoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy, Safety, and Tokenscited by 0
  10. 2026Multi-Dimensional Comparative Scale Construction for Efficient Personalized Subjective Judgment in High-Traffic Applicationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!
  2. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment
  3. 2025Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency
  4. 2025SafeLawBench: Towards Safe Alignment of Large Language Models
  5. 2025Personalized Safety in LLMs: A Benchmark and A Planning-Based Agent Approach
  6. 2025SafetyFlow: An Agent-Flow System for Automated LLM Safety Benchmarking
  7. 2023All Languages Matter: On the Multilingual Safety of Large Language Models
  8. 2025LLM-Safety Evaluations Lack Robustness