Image de couverture de livre

Purify Once, Edit Freely: Breaking Image Protections under Model Mismatch

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhao, Qichen, Zhai, Shengfang, Bai, Xinjian, Shen, Qingni, Lin, Qiqi, Gao, Yansong, Wu, Zhonghai
Format:	Preprint
Publié:	2026
Sujets:	Cryptography and Security Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Life-Cycle Routing Vulnerabilities of LLM Router
par: Lin, Qiqi, et autres
Publié: (2025)

Membership Inference on Text-to-Image Diffusion Models via Conditional Likelihood Discrepancy
par: Zhai, Shengfang, et autres
Publié: (2024)

Discovering Universal Semantic Triggers for Text-to-Image Synthesis
par: Zhai, Shengfang, et autres
Publié: (2024)

SoK: Understanding zk-SNARKs: The Gap Between Research and Practice
par: Liang, Junkai, et autres
Publié: (2025)

Efficient Input-level Backdoor Defense on Text-to-Image Synthesis via Neuron Activation Variation
par: Zhai, Shengfang, et autres
Publié: (2025)

Securing LLM Agents Need Intent-to-Execution Integrity
par: Qu, Wenjie, et autres
Publié: (2026)

Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends
par: Xu, Zhenhua, et autres
Publié: (2025)

BadDLM: Backdooring Diffusion Language Models with Diverse Targets
par: Zhai, Shengfang, et autres
Publié: (2026)

Purifying Approximate Differential Privacy with Randomized Post-processing
par: Lin, Yingyu, et autres
Publié: (2025)

A Survey on Privacy Risks and Protection in Large Language Models
par: Chen, Kang, et autres
Publié: (2025)

Image Privacy Protection: A Survey
par: Wen, Wenying, et autres
Publié: (2024)

TrojanEdit: Multimodal Backdoor Attack Against Image Editing Model
par: Guo, Ji, et autres
Publié: (2024)

Layer-Aware Representation Filtering: Purifying Finetuning Data to Preserve LLM Safety Alignment
par: Li, Hao, et autres
Publié: (2025)

EditShield: Protecting Unauthorized Image Editing by Instruction-guided Diffusion Models
par: Chen, Ruoxi, et autres
Publié: (2023)

Prompt Inference Attack on Distributed Large Language Model Inference Frameworks
par: Luo, Xinjian, et autres
Publié: (2025)

Large Language Model Adversarial Landscape Through the Lens of Attack Objectives
par: Wang, Nan, et autres
Publié: (2025)

DMark: Order-Agnostic Watermarking for Diffusion Large Language Models
par: Wu, Linyu, et autres
Publié: (2025)

Generate "Normal", Edit Poisoned: Branding Injection via Hint Embedding in Image Editing
par: Sun, Desen, et autres
Publié: (2026)

Purified and Unified Steganographic Network
par: Li, Guobiao, et autres
Publié: (2024)

NADD: Amplifying Noise for Effective Diffusion-based Adversarial Purification
par: Nguyen, David D., et autres
Publié: (2026)

Class-Conditional Neural Polarizer: A Lightweight and Effective Backdoor Defense by Purifying Poisoned Features
par: Zhu, Mingli, et autres
Publié: (2025)

Attack as Defense: Run-time Backdoor Implantation for Image Content Protection
par: Zhang, Haichuan, et autres
Publié: (2024)

CodePurify: Defend Backdoor Attacks on Neural Code Models via Entropy-based Purification
par: Mu, Fangwen, et autres
Publié: (2024)

Glaze: Protecting Artists from Style Mimicry by Text-to-Image Models
par: Shan, Shawn, et autres
Publié: (2023)

Exploiting Defenses against GAN-Based Feature Inference Attacks in Federated Learning
par: Luo, Xinjian, et autres
Publié: (2020)

Calibrating Noise for Group Privacy in Subsampled Mechanisms
par: Jiang, Yangfan, et autres
Publié: (2024)

gh0stEdit: Exploiting Layer-Based Access Vulnerability Within Docker Container Images
par: Mills, Alan, et autres
Publié: (2025)

Comprehensive Evaluation of Cloaking Backdoor Attacks on Object Detector in Real-World
par: Ma, Hua, et autres
Publié: (2025)

Intellectual Property Protection for Deep Learning Model and Dataset Intelligence
par: Jiang, Yongqi, et autres
Publié: (2024)

Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems through Benign Queries
par: Wang, Yuhao, et autres
Publié: (2025)

Breaking Free: Efficient Multi-Party Private Set Union Without Non-Collusion Assumptions
par: Dong, Minglang, et autres
Publié: (2024)

Protecting Personalized Trajectory with Differential Privacy under Temporal Correlations
par: Cao, Mingge, et autres
Publié: (2024)

Embedding Integer Lattices as Ideals into Polynomial Rings
par: Cheng, Yihang, et autres
Publié: (2023)

Generalized Implicit Factorization Problem
par: Feng, Yansong, et autres
Publié: (2023)

"No Matter What You Do": Purifying GNN Models via Backdoor Unlearning
par: Zhang, Jiale, et autres
Publié: (2024)

One-to-Multiple Clean-Label Image Camouflage (OmClic) based Backdoor Attack on Deep Learning
par: Wang, Guohong, et autres
Publié: (2023)

DeepiSign-G: Generic Watermark to Stamp Hidden DNN Parameters for Self-contained Tracking
par: Abuadbba, Alsharif, et autres
Publié: (2024)

A Login Page Transparency and Visual Similarity Based Zero Day Phishing Defense Protocol
par: Varshney, Gaurav, et autres
Publié: (2025)

IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation
par: Guo, Yanpei, et autres
Publié: (2026)

High-Rate Public-Key Pseudorandom Codes for Edit Errors
par: Huang, Shengtang, et autres
Publié: (2026)