Buchumschlag

VLM-Guard: Safeguarding Vision-Language Models via Fulfilling Safety Alignment Gap

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Liu, Qin, Wang, Fei, Xiao, Chaowei, Chen, Muhao
Format:	Preprint
Veröffentlicht:	2025
Schlagworte:	Cryptography and Security Artificial Intelligence Computer Vision and Pattern Recognition
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shield Prompting
von: Wang, Yu, et al.
Veröffentlicht: (2024)

FaceSwapGuard: Safeguarding Facial Privacy from DeepFake Threats through Identity Obfuscation
von: Wang, Li, et al.
Veröffentlicht: (2025)

GuardTrace-VL: Detecting Unsafe Multimodel Reasoning via Iterative Safety Supervision
von: Xiang, Yuxiao, et al.
Veröffentlicht: (2025)

SafeVision: Efficient Image Guardrail with Robust Policy Adherence and Explainability
von: Xu, Peiyang, et al.
Veröffentlicht: (2025)

Jailbreaking Safeguarded Text-to-Image Models via Large Language Models
von: Jiang, Zhengyuan, et al.
Veröffentlicht: (2025)

IdentityGuard: Context-Aware Restriction and Provenance for Personalized Synthesis
von: Zhang, Lingyun, et al.
Veröffentlicht: (2026)

PRISM: Robust VLM Alignment with Principled Reasoning for Integrated Safety in Multimodality
von: Li, Nanxi, et al.
Veröffentlicht: (2025)

PromptGuard: Soft Prompt-Guided Unsafe Content Moderation for Text-to-Image Models
von: Yuan, Lingzhi, et al.
Veröffentlicht: (2025)

OmniGuard: Unified Omni-Modal Guardrails with Deliberate Reasoning
von: Zhu, Boyu, et al.
Veröffentlicht: (2025)

Text is All You Need for Vision-Language Model Jailbreaking
von: Chen, Yihang, et al.
Veröffentlicht: (2026)

Disrupting Vision-Language Model-Driven Navigation Services via Adversarial Object Fusion
von: Xie, Chunlong, et al.
Veröffentlicht: (2025)

HomeSafe-Bench: Evaluating Vision-Language Models on Unsafe Action Detection for Embodied Agents in Household Scenarios
von: Pu, Jiayue, et al.
Veröffentlicht: (2026)

On the Adversarial Robustness of Large Vision-Language Models under Visual Token Compression
von: Zhang, Xinwei, et al.
Veröffentlicht: (2026)

The Safety Reminder: A Soft Prompt to Reactivate Delayed Safety Awareness in Vision-Language Models
von: Tang, Peiyuan, et al.
Veröffentlicht: (2025)

Who Can See Through You? Adversarial Shielding Against VLM-Based Attribute Inference Attacks
von: Fan, Yucheng, et al.
Veröffentlicht: (2025)

DAVSP: Safety Alignment for Large Vision-Language Models via Deep Aligned Visual Safety Prompt
von: Zhang, Yitong, et al.
Veröffentlicht: (2025)

Learning to Detect Unseen Jailbreak Attacks in Large Vision-Language Models
von: Liang, Shuang, et al.
Veröffentlicht: (2025)

AGMark: Attention-Guided Dynamic Watermarking for Large Vision-Language Models
von: Li, Yue, et al.
Veröffentlicht: (2026)

From Shortcuts to Triggers: Backdoor Defense with Denoised PoE
von: Liu, Qin, et al.
Veröffentlicht: (2023)

T2I-RiskyPrompt: A Benchmark for Safety Evaluation, Attack, and Defense on Text-to-Image Model
von: Zhang, Chenyu, et al.
Veröffentlicht: (2025)

Beyond the Safety Tax: Mitigating Unsafe Text-to-Image Generation via External Safety Rectification
von: Meng, Xiangtao, et al.
Veröffentlicht: (2025)

AttackVLA: Benchmarking Adversarial and Backdoor Attacks on Vision-Language-Action Models
von: Li, Jiayu, et al.
Veröffentlicht: (2025)

FreezeAsGuard: Mitigating Illegal Adaptation of Diffusion Models via Selective Tensor Freezing
von: Huang, Kai, et al.
Veröffentlicht: (2024)

Spot Risks Before Speaking! Unraveling Safety Attention Heads in Large Vision-Language Models
von: Zheng, Ziwei, et al.
Veröffentlicht: (2025)

Beyond Known Fakes: Generalized Detection of AI-Generated Images via Post-hoc Distribution Alignment
von: Wang, Li, et al.
Veröffentlicht: (2025)

Bodhi VLM: Privacy-Alignment Modeling for Hierarchical Visual Representations in Vision Backbones and VLM Encoders via Bottom-Up and Top-Down Feature Search
von: Ma, Bo, et al.
Veröffentlicht: (2026)

Are Neuro-Inspired Multi-Modal Vision-Language Models Resilient to Membership Inference Privacy Leakage?
von: Amebley, David, et al.
Veröffentlicht: (2025)

Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
von: Ma, Xingjun, et al.
Veröffentlicht: (2025)

Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
von: Li, Xiao, et al.
Veröffentlicht: (2026)

PII-VisBench: Evaluating Personally Identifiable Information Safety in Vision Language Models Along a Continuum of Visibility
von: Shahariar, G M, et al.
Veröffentlicht: (2026)

Adversarial Robustness of Vision in Open Foundation Models
von: Fox, Jonathon, et al.
Veröffentlicht: (2025)

Making Every Step Effective: Jailbreaking Large Vision-Language Models Through Hierarchical KV Equalization
von: Hao, Shuyang, et al.
Veröffentlicht: (2025)

SPQR: A Standardized Benchmark for Modern Safety Alignment Methods in Text-to-Image Diffusion Models
von: Alam, Mohammed Talha, et al.
Veröffentlicht: (2025)

EO-VLM: VLM-Guided Energy Overload Attacks on Vision Models
von: Seo, Minjae, et al.
Veröffentlicht: (2025)

The Structural Safety Generalization Problem
von: Broomfield, Julius, et al.
Veröffentlicht: (2025)

PLA: Prompt Learning Attack against Text-to-Image Generative Models
von: Lyu, Xinqi, et al.
Veröffentlicht: (2025)

Backdoor Attack Against Vision Transformers via Attention Gradient-Based Image Erosion
von: Guo, Ji, et al.
Veröffentlicht: (2024)

DREAM: Scalable Red Teaming for Text-to-Image Generative Systems via Distribution Modeling
von: Li, Boheng, et al.
Veröffentlicht: (2025)

GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning
von: Liu, Yue, et al.
Veröffentlicht: (2025)

Fight Perturbations with Perturbations: Defending Adversarial Attacks via Neuron Influence
von: Chen, Ruoxi, et al.
Veröffentlicht: (2021)