Contrastive Privacy: A Semantic Approach to Measuring Privacy of AI-based Sanitization

arXiv:2605.02977 · cs.CR · Submitted 2026-05-03 · Read on arXiv

cs.CR

Submitted: 2026-05-03

Updated: 2026-09-16

Code: https://github.com/umass-forensics/contrastive-privacy

Project page: https://umass-forensics.github.io/contrastive-privacy

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Differential Privacy — 13× in this paper · explained in 4 episodes
• Cosine Similarity — 6× in this paper · explained in 3 episodes
• Fine-Tuning — 4× in this paper · explained in 5 episodes
• Semantic Similarity — 4× in this paper · explained in 2 episodes
• Actor — 3× in this paper · explained in 3 episodes
• Contextual Integrity — 3× in this paper
• Differential Privacy (DP) — 2× in this paper · explained in 7 episodes
• Differentially Private — 2× in this paper · explained in 3 episodes
• Diffusion Model — 2× in this paper · explained in 8 episodes
• Diffusion Models — 2× in this paper · explained in 6 episodes
• Embeddings — 2× in this paper · explained in 2 episodes
• Generative Model — 2× in this paper · explained in 2 episodes
• Generative Models — 2× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Accountability — 1× in this paper · explained in 2 episodes
• Adversarial Training — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Core — 1× in this paper · explained in 2 episodes
• Latent Space — 1× in this paper · explained in 7 episodes
• Network and Distributed System Security — 1× in this paper
• Pruning — 1× in this paper · explained in 4 episodes

Sources

• From Pixels to Privacy: Temporally Consistent Video Anonymization via Token Pruning for Privacy Preserving Action Recognition
• Unsafe2Safe: Controllable Image Anonymization for Downstream Utility
• The Faiss library
• Beyond Anonymization: Object Scrubbing for Privacy-Preserving 2D and 3D Vision Tasks
• Persuasive Privacy
• Can Large Language Models Really Recognize Your Name?
• GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents
• A False Sense of Privacy: Evaluating Textual Data Sanitization Beyond Surface-level Privacy Leakage
• DP-Image: Differential Privacy for Image Data in Feature Space
• Context-Aware Full Body Anonymization using Text-to-Image Diffusion Models

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