Buchumschlag

When Surfaces Lie: Exploiting Wrinkle-Induced Attention Shift to Attack Vision-Language Models

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Hu, Chengyin, Sun, Xuemeng, Han, Jiaju, Zhang, Qike, Chen, Xiang, Wang, Xin, Wei, Yiwei, Long, Jiahua
Format:	Preprint
Veröffentlicht:	2026
Schlagworte:	Computer Vision and Pattern Recognition
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

XSPA: Crafting Imperceptible X-Shaped Sparse Adversarial Perturbations for Transferable Attacks on VLMs
von: Hu, Chengyin, et al.
Veröffentlicht: (2026)

From Clouds to Hallucinations: Atmospheric Retrieval Hijacking in Remote Sensing Vision-Language RAG
von: Han, Jiaju, et al.
Veröffentlicht: (2026)

Challenging Vision-Language Models with Physically Deployable Multimodal Semantic Lighting Attacks
von: Zhao, Yingying, et al.
Veröffentlicht: (2026)

Thermal Topology Collapse: Universal Physical Patch Attacks on Infrared Vision Systems
von: Hu, Chengyin, et al.
Veröffentlicht: (2026)

A Semantic Decoupling-Based Two-Stage Rainy-Day Attack for Revealing Weather Robustness Deficiencies in Vision-Language Models
von: Hu, Chengyin, et al.
Veröffentlicht: (2026)

CoDA: Exploring Chain-of-Distribution Attacks and Post-Hoc Token-Space Repair for Medical Vision-Language Models
von: Chen, Xiang, et al.
Veröffentlicht: (2026)

Exposing Vulnerabilities in Visible-Infrared VLMs: A Unified Geometric Adversarial Framework with Cross-Task Transferability
von: Chen, Xiang, et al.
Veröffentlicht: (2026)

Revealing Physical-World Semantic Vulnerabilities: Universal Adversarial Patches for Infrared Vision-Language Models
von: Hu, Chengyin, et al.
Veröffentlicht: (2026)

When and Where to Attack? Stage-wise Attention-Guided Adversarial Attack on Large Vision Language Models
von: Kwak, Jaehyun, et al.
Veröffentlicht: (2026)

Multi-View Black-Box Physical Attacks on Infrared Pedestrian Detectors Using Adversarial Infrared Grid
von: Tiliwalidi, Kalibinuer, et al.
Veröffentlicht: (2024)

Fairness-aware Vision Transformer via Debiased Self-Attention
von: Qiang, Yao, et al.
Veröffentlicht: (2023)

QuantAttack: Exploiting Dynamic Quantization to Attack Vision Transformers
von: Baras, Amit, et al.
Veröffentlicht: (2023)

Attention-Guided Patch-Wise Sparse Adversarial Attacks on Vision-Language-Action Models
von: Zhang, Naifu, et al.
Veröffentlicht: (2025)

Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models
von: Wang, Hongjun, et al.
Veröffentlicht: (2026)

Interpretability-Aware Vision Transformer
von: Qiang, Yao, et al.
Veröffentlicht: (2023)

OFFSET: Segmentation-based Focus Shift Revision for Composed Image Retrieval
von: Chen, Zhiwei, et al.
Veröffentlicht: (2025)

CILP-FGDI: Exploiting Vision-Language Model for Generalizable Person Re-Identification
von: Zhao, Huazhong, et al.
Veröffentlicht: (2025)

When Text Hijacks Vision: Benchmarking and Mitigating Text Overlay-Induced Hallucination in Vision Language Models
von: Yakun, Cui, et al.
Veröffentlicht: (2026)

Ensuring Force Safety in Vision-Guided Robotic Manipulation via Implicit Tactile Calibration
von: Wei, Lai, et al.
Veröffentlicht: (2024)

When Lighting Deceives: Exposing Vision-Language Models' Illumination Vulnerability Through Illumination Transformation Attack
von: Liu, Hanqing, et al.
Veröffentlicht: (2025)

When Background Matters: Breaking Medical Vision Language Models by Transferable Attack
von: Ghosh, Akash, et al.
Veröffentlicht: (2026)

Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift
von: Liang, Siyuan, et al.
Veröffentlicht: (2024)

Interpretable Modeling of Driver Attention Shifts with a Vision--Language Model
von: Hamid, Kaiser, et al.
Veröffentlicht: (2025)

BiPVL-Seg: Bidirectional Progressive Vision-Language Fusion with Global-Local Alignment for Medical Image Segmentation
von: Sultan, Rafi Ibn, et al.
Veröffentlicht: (2025)

Optimizing Vision-Language Consistency via Cross-Layer Regional Attention Alignment
von: Wang, Yifan, et al.
Veröffentlicht: (2025)

Towards Zero-Shot Annotation of the Built Environment with Vision-Language Models (Vision Paper)
von: Han, Bin, et al.
Veröffentlicht: (2024)

When Alignment Fails: Multimodal Adversarial Attacks on Vision-Language-Action Models
von: Yan, Yuping, et al.
Veröffentlicht: (2025)

Physical Prompt Injection Attacks on Large Vision-Language Models
von: Ling, Chen, et al.
Veröffentlicht: (2026)

SoLA-Vision: Fine-grained Layer-wise Linear Softmax Hybrid Attention
von: Li, Ruibang, et al.
Veröffentlicht: (2026)

When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models
von: Hou, Jiacheng, et al.
Veröffentlicht: (2026)

Learning to Remove Wrinkled Transparent Film with Polarized Prior
von: Tang, Jiaqi, et al.
Veröffentlicht: (2024)

Mitigating Hallucination in Large Vision-Language Models through Aligning Attention Distribution to Information Flow
von: Zhao, Jianfei, et al.
Veröffentlicht: (2025)

One Object, Multiple Lies: A Benchmark for Cross-task Adversarial Attack on Unified Vision-Language Models
von: Zhao, Jiale, et al.
Veröffentlicht: (2025)

Exploiting Information Redundancy in Attention Maps for Extreme Quantization of Vision Transformers
von: Maisonnave, Lucas, et al.
Veröffentlicht: (2025)

When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language Models
von: Choi, Jiho, et al.
Veröffentlicht: (2026)

Rethinking Causal Mask Attention for Vision-Language Inference
von: Pei, Xiaohuan, et al.
Veröffentlicht: (2025)

When Looking Is Not Enough: Visual Attention Structure Reveals Hallucination in MLLMs
von: Cao, Fanpu, et al.
Veröffentlicht: (2026)

When Robots Obey the Patch: Universal Transferable Patch Attacks on Vision-Language-Action Models
von: Lu, Hui, et al.
Veröffentlicht: (2025)

A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trends
von: Liu, Daizong, et al.
Veröffentlicht: (2024)

RESTORE: Towards Feature Shift for Vision-Language Prompt Learning
von: Yang, Yuncheng, et al.
Veröffentlicht: (2024)