User Perceptions vs. Proxy LLM Judges: Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are rapidly being adopted for tasks like drafting emails, summarizing meetings, and answering health questions. |
| Approach: | They conducted a scenario-based evaluation of Large language models (LLMs) using 90 PrivacyLens scenarios. |
| Outcome: | The proposed models can leak private information in complex scenarios, but they do not measure user perceptions directly. |
Similar Papers
Controlling What You Share: Assessing Language Model Adherence to Privacy Preferences (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are accessed via commercial APIs, but expose data to service providers. |
| Approach: | They propose a framework where a local model uses natural language instructions to rewrite queries and paired them with synthetic privacy profiles to achieve better privacy preservation. |
| Outcome: | The proposed model outperforms large-scale few-shot models in terms of privacy preservation and performance. |
LLMs are Privacy Erasable (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a new study examines the privacy of large language models and their capabilities . the study aims to address the balance between the convenience of LLMs and user privacy concerns . |
| Approach: | They propose a strategy that safeguards user prompt while accessing LLM cloud services . they evaluate the efficacy of their method across prominent LLM benchmarks . |
| Outcome: | The proposed method thwarts reconstruction attacks and improves model performance . it also surpasses the results reported in official model cards . |
Can LLM be a Personalized Judge? (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a new study examines the reliability of large language models (LLMs) for personalization and role-playing evaluation without examining its validity. |
| Approach: | They investigate the reliability of LLM-as-a-Personalized-Judge for personalization . they find that personas provided to LLMs have limited predictive power . |
| Outcome: | The proposed model is less reliable than previously thought, the authors show . human annotation reveals that third-person crowd worker evaluations of personalized preferences are even worse than LLM predictions. |
Can a Large Language Model Keep My Secrets? A Study on LLM-Controlled Agents (2025.acl-srw)
Copied to clipboard
| Challenge: | Using large language models, agents can assist with natural language tasks when given access to confidential data. |
| Approach: | They created a synthetic dataset consisting of confidentiality-aware planning and deduction tasks in organizational access control. |
| Outcome: | The proposed model can perform tasks similar to humans when given access to confidential data. |
Justice in Judgment: Unveiling (Hidden) Bias in LLM-assisted Peer Reviews (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing studies show that large language models carry implicit biases across race, gender, and religion . prior studies documented such biase based on text generation and classification tasks . |
| Approach: | They investigate bias in large language models by controlling metadata on author metadata . authors found affiliation bias favoring authors from highly ranked institutions . |
| Outcome: | The proposed model favors authors from highly ranked institutions, the authors show . the model also favors author affiliations from highly-ranked institutions . |
SecureSQL: Evaluating Data Leakage of Large Language Models as Natural Language Interfaces to Databases (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies on the vulnerability of large language models to SQL injection have been limited. |
| Approach: | They propose to evaluate the potential of language models to leak sensitive data when generating SQL queries. |
| Outcome: | The proposed model with the best performance has an accuracy of 61.7%, compared to humans who achieve 94% accuracy. |
When Can We Trust LLMs in Mental Health? Large-Scale Benchmarks for Reliable LLM Evaluation (2026.eacl-long)
Copied to clipboard
Abeer Badawi, Elahe Rahimi, Md Tahmid Rahman Laskar, Sheri Grach, Lindsay Bertrand, Lames Danok, Prathiba Dhanesh, Jimmy Huang, Frank Rudzicz, Elham Dolatabadi
| Challenge: | Existing benchmarks for large language models are limited in scale, authenticity, and reliability due to the emotionally complex nature of therapeutic dialogue. |
| Approach: | They propose two benchmarks that provide a framework for evaluating large language models for mental health support. |
| Outcome: | The proposed framework provides a framework for generation and evaluation of large-scale authentic dialogue datasets and judge-reliability assessments. |
PII-Bench: Evaluating Query-Aware Privacy Protection Systems (2026.acl-long)
Copied to clipboard
| Challenge: | Existing models do not detect PII in user prompts, despite their convenience . current models show significant limitations in determining PI I query relevance . |
| Approach: | They propose a query-unrelated PII masking strategy and propose PIi-Bench . they propose 'quick-and-easy' PI I masking with a user query and context description . |
| Outcome: | The proposed model performs well in basic PII detection, but shows significant limitations in query relevance. |
Value–Action Alignment in Large Language Models under Privacy–Prosocial Conflict (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluations measure privacy-related attitudes or sharing intentions in isolation, making it difficult to determine whether a model’s expressed values jointly predict its downstream data-sharing actions as in real human behaviors. |
| Approach: | They propose a framework that sequentially administers standardized questionnaires for privacy attitudes, prosocialness, and acceptance of data sharing within a bounded, history-carrying session. |
| Outcome: | The proposed model shows that it is stable but model-specific Privacy-PSA-AoDS profiles, and substantial heterogeneity in value-action alignment. |
Humans or LLMs as the Judge? A Study on Judgement Bias (2024.emnlp-main)
Copied to clipboard
| Challenge: | Proprietary models such as GPT-4, Claude, Gemini-Pro and others are being democratized to improve evaluations of LLMs. |
| Approach: | They propose a framework that is free from referencing groundtruth annotations for investigating **Misinformation Oversight Bias**, **Gender Bia**,**Authority Bia* and **Beauty Bia's** on LLM and human judges. |
| Outcome: | The proposed framework investigates **Misinformation Oversight Bias**, **Gender Bia**,**Authority Bia* and **Beauty Bia' on LLM and human judges. |