Image de couverture de livre

SafeRedir: Prompt Embedding Redirection for Robust Unlearning in Image Generation Models

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Liu, Renyang, Chen, Kangjie, Qiu, Han, Zhang, Jie, Lam, Kwok-Yan, Zhang, Tianwei, Ng, See-Kiong
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence Cryptography and Security Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Image Can Bring Your Memory Back: A Novel Multi-Modal Guided Attack against Image Generation Model Unlearning
par: Liu, Renyang, et autres
Publié: (2025)

Rethinking Machine Unlearning in Image Generation Models
par: Liu, Renyang, et autres
Publié: (2025)

DECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systems
par: Ou, Haoran, et autres
Publié: (2026)

Towards Effective Prompt Stealing Attack against Text-to-Image Diffusion Models
par: Zhao, Shiqian, et autres
Publié: (2025)

ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users
par: Li, Guanlin, et autres
Publié: (2024)

Safe Text-to-Image Generation: Simply Sanitize the Prompt Embedding
par: Qiu, Huming, et autres
Publié: (2024)

When Search Goes Wrong: Red-Teaming Web-Augmented Large Language Models
par: Ou, Haoran, et autres
Publié: (2025)

CAAP: Capture-Aware Adversarial Patch Attacks on Palmprint Recognition Models
par: Liu, Renyang, et autres
Publié: (2026)

ObfusBFA: A Holistic Approach to Safeguarding DNNs from Different Types of Bit-Flip Attacks
par: Yan, Xiaobei, et autres
Publié: (2025)

Robust-Wide: Robust Watermarking against Instruction-driven Image Editing
par: Hu, Runyi, et autres
Publié: (2024)

Guaranteeing Data Privacy in Federated Unlearning with Dynamic User Participation
par: Liu, Ziyao, et autres
Publié: (2024)

Privacy-Preserving Federated Unlearning with Certified Client Removal
par: Liu, Ziyao, et autres
Publié: (2024)

A Survey on Federated Unlearning: Challenges, Methods, and Future Directions
par: Liu, Ziyao, et autres
Publié: (2023)

BURN: Backdoor Unlearning via Adversarial Boundary Analysis
par: Su, Yanghao, et autres
Publié: (2025)

Threats, Attacks, and Defenses in Machine Unlearning: A Survey
par: Liu, Ziyao, et autres
Publié: (2024)

Confidence Elicitation: A New Attack Vector for Large Language Models
par: Formento, Brian, et autres
Publié: (2025)

Certifying the Right to Be Forgotten: Primal-Dual Optimization for Sample and Label Unlearning in Vertical Federated Learning
par: Jiang, Yu, et autres
Publié: (2025)

Towards Action Hijacking of Large Language Model-based Agent
par: Zhang, Yuyang, et autres
Publié: (2024)

Efficient Federated Unlearning with Adaptive Differential Privacy Preservation
par: Jiang, Yu, et autres
Publié: (2024)

The Trojan Example: Jailbreaking LLMs through Template Filling and Unsafety Reasoning
par: Liu, Mingrui, et autres
Publié: (2025)

SafeRedirect: Defeating Internal Safety Collapse via Task-Completion Redirection in Frontier LLMs
par: Pan, Chao, et autres
Publié: (2026)

RedVisor: Reasoning-Aware Prompt Injection Defense via Zero-Copy KV Cache Reuse
par: Liu, Mingrui, et autres
Publié: (2026)

Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model
par: Wu, Tianyi, et autres
Publié: (2026)

Geneshift: Impact of different scenario shift on Jailbreaking LLM
par: Wu, Tianyi, et autres
Publié: (2025)

An Engorgio Prompt Makes Large Language Model Babble on
par: Dong, Jianshuo, et autres
Publié: (2024)

BitHydra: Towards Bit-flip Inference Cost Attack against Large Language Models
par: Yan, Xiaobei, et autres
Publié: (2025)

SafeGuider: Robust and Practical Content Safety Control for Text-to-Image Models
par: Qi, Peigui, et autres
Publié: (2025)

PoseGuard: Pose-Guided Generation with Safety Guardrails
par: Wang, Kongxin, et autres
Publié: (2025)

LLMs Cannot Reliably Judge (Yet?): A Comprehensive Assessment on the Robustness of LLM-as-a-Judge
par: Li, Songze, et autres
Publié: (2025)

Efficient Privacy-Preserving Retrieval Augmented Generation with Distance-Preserving Encryption
par: Ye, Huanyi, et autres
Publié: (2026)

REFORGE: Multi-modal Attacks Reveal Vulnerable Concept Unlearning in Image Generation Models
par: Zou, Yong, et autres
Publié: (2026)

State-Dependent Safety Failures in Multi-Turn Language Model Interaction
par: Li, Pengcheng, et autres
Publié: (2026)

BadLingual: A Novel Lingual-Backdoor Attack against Large Language Models
par: Wang, Zihan, et autres
Publié: (2025)

BadEdit: Backdooring large language models by model editing
par: Li, Yanzhou, et autres
Publié: (2024)

Robust Reversible Watermarking in Encrypted Images Based on Dual-MSBs Spiral Embedding
par: Shen, Haoyu, et autres
Publié: (2026)

ForgetMark: Stealthy Fingerprint Embedding via Targeted Unlearning in Language Models
par: Xu, Zhenhua, et autres
Publié: (2026)

DeepSweep: An Evaluation Framework for Mitigating DNN Backdoor Attacks using Data Augmentation
par: Qiu, Han, et autres
Publié: (2020)

On the Proactive Generation of Unsafe Images From Text-To-Image Models Using Benign Prompts
par: Wu, Yixin, et autres
Publié: (2023)

Really Unlearned? Verifying Machine Unlearning via Influential Sample Pairs
par: Xu, Heng, et autres
Publié: (2024)

What Makes a Good LLM Agent for Real-world Penetration Testing?
par: Deng, Gelei, et autres
Publié: (2026)