Imagen de Portada

TAIJI: Textual Anchoring for Immunizing Jailbreak Images in Vision Language Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Yin, Xiangyu, Qi, Yi, Hu, Jinwei, Chen, Zhen, Dong, Yi, Zhao, Xingyu, Huang, Xiaowei, Ruan, Wenjie
Formato:	Preprint
Publicado:	2025
Materias:	Computer Vision and Pattern Recognition Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

CeTAD: Towards Certified Toxicity-Aware Distance in Vision Language Models
por: Yin, Xiangyu, et al.
Publicado: (2025)

Fragile by Design: On the Limits of Adversarial Defenses in Personalized Generation
por: Chen, Zhen, et al.
Publicado: (2025)

ProTIP: Probabilistic Robustness Verification on Text-to-Image Diffusion Models against Stochastic Perturbation
por: Zhang, Yi, et al.
Publicado: (2024)

GPO-V: Jailbreak Diffusion Vision Language Model by Global Probability Optimization
por: Pan, Yu, et al.
Publicado: (2026)

Boosting Adversarial Training via Fisher-Rao Norm-based Regularization
por: Yin, Xiangyu, et al.
Publicado: (2024)

Activation Steering Meets Preference Optimization: Defense Against Jailbreaks in Vision Language Models
por: Wu, Sihao, et al.
Publicado: (2025)

FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model
por: Hu, Jinwei, et al.
Publicado: (2025)

ImgTrojan: Jailbreaking Vision-Language Models with ONE Image
por: Tao, Xijia, et al.
Publicado: (2024)

A Black-Box Evaluation Framework for Semantic Robustness in Bird's Eye View Detection
por: Wang, Fu, et al.
Publicado: (2024)

SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model
por: Huang, Zhenglin, et al.
Publicado: (2024)

Principled Steering via Null-space Projection for Jailbreak Defense in Vision-Language Models
por: Zhu, Xingyu, et al.
Publicado: (2026)

Building Guardrails for Large Language Models
por: Dong, Yi, et al.
Publicado: (2024)

Spatial-DISE: A Unified Benchmark for Evaluating Spatial Reasoning in Vision-Language Models
por: Huang, Xinmiao, et al.
Publicado: (2025)

Cross-Modal Obfuscation for Jailbreak Attacks on Large Vision-Language Models
por: Jiang, Lei, et al.
Publicado: (2025)

Risk Controlled Image Retrieval
por: Cai, Kaiwen, et al.
Publicado: (2023)

Contrastive Language-Image Learning with Augmented Textual Prompts for 3D/4D FER Using Vision-Language Model
por: Behzad, Muzammil, et al.
Publicado: (2025)

PRBench: A Standardized Probabilistic Robustness Benchmark
por: Zhang, Yi, et al.
Publicado: (2025)

Proxy Robustness in Vision Language Models is Effortlessly Transferable
por: Fu, Xiaowei, et al.
Publicado: (2026)

Advancing Textual Prompt Learning with Anchored Attributes
por: Li, Zheng, et al.
Publicado: (2024)

Exploring Textual Semantics Diversity for Image Transmission in Semantic Communication Systems using Visual Language Model
por: Huang, Peishan, et al.
Publicado: (2025)

JaiLIP: Jailbreaking Vision-Language Models via Loss Guided Image Perturbation
por: Mia, Md Jueal, et al.
Publicado: (2025)

Immuno-VLM: Immunizing Large Vision-Language Models via Generative Semantic Antibodies for Open-World Trustworthiness
por: Fang, Xiang, et al.
Publicado: (2026)

Towards Real-World Adverse Weather Image Restoration: Enhancing Clearness and Semantics with Vision-Language Models
por: Xu, Jiaqi, et al.
Publicado: (2024)

Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models
por: Zeng, Yu, et al.
Publicado: (2026)

Towards Fairness-Aware Adversarial Learning
por: Zhang, Yanghao, et al.
Publicado: (2024)

JailbreakZoo: Survey, Landscapes, and Horizons in Jailbreaking Large Language and Vision-Language Models
por: Jin, Haibo, et al.
Publicado: (2024)

Towards Long-window Anchoring in Vision-Language Model Distillation
por: Zhou, Haoyi, et al.
Publicado: (2025)

ImageRef-VL: Enabling Contextual Image Referencing in Vision-Language Models
por: Yi, Jingwei, et al.
Publicado: (2025)

Enhancing Vision-Language Tracking by Effectively Converting Textual Cues into Visual Cues
por: Feng, X., et al.
Publicado: (2024)

Jailbreaks on Vision Language Model via Multimodal Reasoning
por: Noheria, Aarush, et al.
Publicado: (2026)

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

Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models
por: Ruan, Shouwei, et al.
Publicado: (2024)

Immunizing Images from Text to Image Editing via Adversarial Cross-Attention
por: Trippodo, Matteo, et al.
Publicado: (2025)

ViTCoP: Accelerating Large Vision-Language Models via Visual and Textual Semantic Collaborative Pruning
por: Luo, Wen, et al.
Publicado: (2026)

Vision Foundation Models as Generalist Tokenizers for Image Generation
por: Zheng, Anlin, et al.
Publicado: (2026)

Jailbreak Large Vision-Language Models Through Multi-Modal Linkage
por: Wang, Yu, et al.
Publicado: (2024)

JailBound: Jailbreaking Internal Safety Boundaries of Vision-Language Models
por: Song, Jiaxin, et al.
Publicado: (2025)

Trust-Oriented Adaptive Guardrails for Large Language Models
por: Hu, Jinwei, et al.
Publicado: (2024)

Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models
por: Cui, Kaiyuan, et al.
Publicado: (2026)

Exploring Visual Vulnerabilities via Multi-Loss Adversarial Search for Jailbreaking Vision-Language Models
por: Hao, Shuyang, et al.
Publicado: (2024)