Image de couverture de livre

Trusted Machine Learning Models Unlock Private Inference for Problems Currently Infeasible with Cryptography

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Shumailov, Ilia, Ramage, Daniel, Meiklejohn, Sarah, Kairouz, Peter, Hartmann, Florian, Balle, Borja, Bagdasarian, Eugene
Format:	Preprint
Publié:	2025
Sujets:	Cryptography and Security Artificial Intelligence Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

AirGapAgent: Protecting Privacy-Conscious Conversational Agents
par: Bagdasarian, Eugene, et autres
Publié: (2024)

Machine Learning Models Have a Supply Chain Problem
par: Meiklejohn, Sarah, et autres
Publié: (2025)

Beyond Laplace and Gaussian: Exploring the Generalized Gaussian Mechanism for Private Machine Learning
par: Rinberg, Roy, et autres
Publié: (2025)

Can Large Language Models Really Recognize Your Name?
par: Pham, Dzung, et autres
Publié: (2025)

Machine Learning needs Better Randomness Standards: Randomised Smoothing and PRNG-based attacks
par: Dahiya, Pranav, et autres
Publié: (2023)

CLIOPATRA: Extracting Private Information from LLM Insights
par: Annamalai, Meenatchi Sundaram Muthu Selva, et autres
Publié: (2026)

Behavioral Canaries: Auditing Private Retrieved Context Usage in RL Fine-Tuning
par: Chen, Chaoran, et autres
Publié: (2026)

Kraken: Higher-order EM Side-Channel Attacks on DNNs in Near and Far Field
par: Horvath, Peter, et autres
Publié: (2026)

Contrastive Privacy: A Semantic Approach to Measuring Privacy of AI-based Sanitization
par: Bissias, George, et autres
Publié: (2026)

SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks
par: Gao, Yue, et autres
Publié: (2023)

Buffer Overflow in Mixture of Experts
par: Hayes, Jamie, et autres
Publié: (2024)

Architectural Backdoors for Within-Batch Data Stealing and Model Inference Manipulation
par: Küchler, Nicolas, et autres
Publié: (2025)

DiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction
par: Zhang, Xinwei, et autres
Publié: (2024)

Contextual Agent Security: A Policy for Every Purpose
par: Tsai, Lillian, et autres
Publié: (2025)

AI Agents May Always Fall for Prompt Injections
par: Abdelnabi, Sahar, et autres
Publié: (2026)

On the Privacy of Selection Mechanisms with Gaussian Noise
par: Lebensold, Jonathan, et autres
Publié: (2024)

Secure Stateful Aggregation: A Practical Protocol with Applications in Differentially-Private Federated Learning
par: Ball, Marshall, et autres
Publié: (2024)

Federated Learning in Practice: Reflections and Projections
par: Daly, Katharine, et autres
Publié: (2024)

ceLLMate: Sandboxing Browser AI Agents
par: Meng, Luoxi, et autres
Publié: (2025)

Practical Secure Aggregation by Combining Cryptography and Trusted Execution Environments
par: de Laage, Romain, et autres
Publié: (2025)

Is API Access to LLMs Useful for Generating Private Synthetic Tabular Data?
par: Swanberg, Marika, et autres
Publié: (2025)

Beyond Labeling Oracles: What does it mean to steal ML models?
par: Shafran, Avital, et autres
Publié: (2023)

Disappearing Ink: Obfuscation Breaks N-gram Code Watermarks in Theory and Practice
par: Zhang, Gehao, et autres
Publié: (2025)

Differentially Private Stream Processing at Scale
par: Zhang, Bing, et autres
Publié: (2023)

Toward provably private analytics and insights into GenAI use
par: Cheu, Albert, et autres
Publié: (2025)

ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks
par: Clifford, Eleanor, et autres
Publié: (2022)

Fixing 7,400 Bugs for 1$: Cheap Crash-Site Program Repair
par: Zheng, Han, et autres
Publié: (2025)

Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy
par: Hayes, Jamie, et autres
Publié: (2024)

When Vision Fails: Text Attacks Against ViT and OCR
par: Boucher, Nicholas, et autres
Publié: (2023)

A Unified Framework for Adversary-Aware Differential Privacy Bounds
par: Swanberg, Marika, et autres
Publié: (2025)

Watermarking Needs Input Repetition Masking
par: Khachaturov, David, et autres
Publié: (2025)

Beyond Slow Signs in High-fidelity Model Extraction
par: Foerster, Hanna, et autres
Publié: (2024)

ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted Control
par: Hu, Yuzheng, et autres
Publié: (2025)

Soft Instruction De-escalation Defense
par: Walter, Nils Philipp, et autres
Publié: (2025)

Network-Level Prompt and Trait Leakage in Local Research Agents
par: Jeong, Hyejun, et autres
Publié: (2025)

Prompt Public Large Language Models to Synthesize Data for Private On-device Applications
par: Wu, Shanshan, et autres
Publié: (2024)

Scaling Laws for Differentially Private Language Models
par: McKenna, Ryan, et autres
Publié: (2025)

Backdooring Bias ($B^2$) into Stable Diffusion Models
par: Naseh, Ali, et autres
Publié: (2024)

The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD
par: Steinke, Thomas, et autres
Publié: (2024)

A Pervasive, Efficient and Private Future: Realizing Privacy-Preserving Machine Learning Through Hybrid Homomorphic Encryption
par: Nguyen, Khoa, et autres
Publié: (2024)