PrivacyGuard: Source-Linked Sanitization and Locally Resolvable PII References for Cloud GUI Agents
cs.CR, cs.AI, cs.DC
Submitted: 2026-10-12
Updated: 2026-10-12
Project page: https://microsoft.github.io/presidio
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- LLM-Redactor: An Empirical Evaluation of Eight Techniques for Privacy-Preserving LLM Requests
- ScreenAI: A Vision-Language Model for UI and Infographics Understanding
- Hide and Seek (HaS): A Lightweight Framework for Prompt Privacy Protection
- SurrogateShield: Beyond Redaction for High-Utility, Privacy-Preserving LLM Interactions
- PIIBench: A Unified Multi-Source Benchmark Corpus for Personally Identifiable Information Detection
- AppAgentX: Evolving GUI Agents as Proficient Smartphone Users
- AppAgent v2: Advanced Agent for Flexible Mobile Interactions
- OmniParser for Pure Vision Based GUI Agent
- Split-and-Denoise: Protect large language model inference with local differential privacy
- DiffSTR: Controlled Diffusion Models for Scene Text Removal
- UI-TARS: Pioneering Automated GUI Interaction with Native Agents
- CAPED: Context-Aware Privacy Exposure Defense for Mobile GUI Agents
- Towards Trustworthy GUI Agents: A Survey
- A System for Name and Address Parsing with Large Language Models
- Mobile-Agent: Autonomous Multi-Modal Mobile Device Agent with Visual Perception
- Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents
- Mobile-Agent-v3: Fundamental Agents for GUI Automation
- Large Language Model-Brained GUI Agents: A Survey
- DualTAP: A Dual-Task Adversarial Protector for Mobile MLLM Agents
- Anonymization-Enhanced Privacy Protection for Mobile GUI Agents: Available but Invisible
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs