# User Perceptions vs. Proxy LLM Judges: Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios

arXiv: https://arxiv.org/abs/2510.20721 · HTML: https://mlsanity.com/paper/2510.20721/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 75 references that are in the map.
- First posted on arXiv 23 Oct 2025.

## Builds on (most cited first, 30 of 37)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [SummEval: Re-evaluating Summarization Evaluation](https://mlsanity.com/paper/2007.12626/) (2020)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory](https://mlsanity.com/paper/2310.17884/) (2023)
- [What Does it Mean for a Language Model to Preserve Privacy?](https://mlsanity.com/paper/2202.05520/) (2022)
- [PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Action](https://mlsanity.com/paper/2409.00138/) (2024)
- [Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning](https://mlsanity.com/paper/2408.10075/) (2024)
- [Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs](https://mlsanity.com/paper/2502.09597/) (2025)
- [Hello Again! LLM-powered Personalized Agent for Long-term Dialogue](https://mlsanity.com/paper/2406.05925/) (2024)
- [Gmail Smart Compose: Real-Time Assisted Writing](https://mlsanity.com/paper/1906.00080/) (2019)
- [AgentDAM: Privacy Leakage Evaluation for Autonomous Web Agents](https://mlsanity.com/paper/2503.09780/) (2025)
- [On the Robustness of Code Generation Techniques: An Empirical Study on GitHub Copilot](https://mlsanity.com/paper/2302.00438/) (2023)
- ["It's a Fair Game", or Is It? Examining How Users Navigate Disclosure Risks and Benefits When Using LLM-Based Conversational Agents](https://mlsanity.com/paper/2309.11653/) (2023)
- [The Obvious Invisible Threat: LLM-Powered GUI Agents' Vulnerability to Fine-Print Injections](https://mlsanity.com/paper/2504.11281/) (2025)
- [BIG5-CHAT: Shaping LLM Personalities Through Training on Human-Grounded Data](https://mlsanity.com/paper/2410.16491/) (2024)
- [Trust No Bot: Discovering Personal Disclosures in Human-LLM Conversations in the Wild](https://mlsanity.com/paper/2407.11438/) (2024)
- [Human-Centered Design Recommendations for LLM-as-a-Judge](https://mlsanity.com/paper/2407.03479/) (2024)
- [PrivaCI-Bench: Evaluating Privacy with Contextual Integrity and Legal Compliance](https://mlsanity.com/paper/2502.17041/) (2025)
- [Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agents](https://mlsanity.com/paper/2509.17488/) (2025)
- [CLEAR: Towards Contextual LLM-Empowered Privacy Policy Analysis and Risk Generation for Large Language Model Applications](https://mlsanity.com/paper/2410.13387/) (2024)
- [Privacy- and Utility-Preserving NLP with Anonymized Data: A case study of Pseudonymization](https://mlsanity.com/paper/2306.05561/) (2023)
- [1-2-3 Check: Enhancing Contextual Privacy in LLM via Multi-Agent Reasoning](https://mlsanity.com/paper/2508.07667/) (2025)
- [One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation](https://mlsanity.com/paper/2402.11683/) (2024)
- [Android Permissions Remystified: A Field Study on Contextual Integrity](https://mlsanity.com/paper/1504.03747/) (2015)
- [LLM Whisperer: An Inconspicuous Attack to Bias LLM Responses](https://mlsanity.com/paper/2406.04755/) (2024)
- [Human-in-the-loop Evaluation for Early Misinformation Detection: A Case Study of COVID-19 Treatments](https://mlsanity.com/paper/2212.09683/) (2022)
- [Privacy Leakage Overshadowed by Views of AI: A Study on Human Oversight of Privacy in Language Model Agent](https://mlsanity.com/paper/2411.01344/) (2024)
- [Summaries, Highlights, and Action items: Design, implementation and evaluation of an LLM-powered meeting recap system](https://mlsanity.com/paper/2307.15793/) (2023)
- [Evaluating Multimodal Language Models as Visual Assistants for Visually Impaired Users](https://mlsanity.com/paper/2503.22610/) (2025)

## Built on by (most cited first, 1 of 1)

- [Say Something Else: Rethinking Contextual Privacy as Information Sufficiency](https://mlsanity.com/paper/2604.06409/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are rapidly being adopted for tasks like drafting emails, summarizing meetings, and answering health questions. In these settings, users may need to share private information (e.g., contact details, health records). To evaluate LLMs' ability to identify and redact such information, prior work introduced real-life, scenario-based benchmarks (e.g., ConfAIde, PrivacyLens) and found that LLMs can leak private information in complex scenarios. However, these evaluations relied on proxy LLMs to judge the helpfulness and privacy-preservation quality of LLM responses, rather than directly measuring users' perceptions. To understand how users perceive the helpfulness and privacy-preservation quality of LLM responses to privacy-sensitive scenarios, we conducted a user study (n=94) using 90 PrivacyLens scenarios. We found that users had low agreement with each other when evaluating identical LLM responses. In contrast, five proxy LLMs reached high agreement, yet each proxy LLM had low correlation with users' evaluations. These results indicate that proxy LLMs cannot accurately estimate users' wide range of perceptions of utility and privacy in privacy-sensitive scenarios. We discuss the need for more user-centered studies to measure LLMs' ability to help users while preserving privacy, and for improving alignment between LLMs and users in estimating perceived privacy and utility.