Image de couverture de livre

Exposing Vulnerabilities in Visible-Infrared VLMs: A Unified Geometric Adversarial Framework with Cross-Task Transferability

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Chen, Xiang, Dong, Yuxian, Li, Chao, Hu, Chengyin, Han, Jiaju, Zhang, Fengyu, Wei, Yiwei, Long, Jiahuan, Guo, Jiujiang
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Revealing Physical-World Semantic Vulnerabilities: Universal Adversarial Patches for Infrared Vision-Language Models
par: Hu, Chengyin, et autres
Publié: (2026)

XSPA: Crafting Imperceptible X-Shaped Sparse Adversarial Perturbations for Transferable Attacks on VLMs
par: Hu, Chengyin, et autres
Publié: (2026)

Thermal Topology Collapse: Universal Physical Patch Attacks on Infrared Vision Systems
par: Hu, Chengyin, et autres
Publié: (2026)

From Clouds to Hallucinations: Atmospheric Retrieval Hijacking in Remote Sensing Vision-Language RAG
par: Han, Jiaju, et autres
Publié: (2026)

CoDA: Exploring Chain-of-Distribution Attacks and Post-Hoc Token-Space Repair for Medical Vision-Language Models
par: Chen, Xiang, et autres
Publié: (2026)

A Semantic Decoupling-Based Two-Stage Rainy-Day Attack for Revealing Weather Robustness Deficiencies in Vision-Language Models
par: Hu, Chengyin, et autres
Publié: (2026)

CDUPatch: Color-Driven Universal Adversarial Patch Attack for Dual-Modal Visible-Infrared Detectors
par: Long, Jiahuan, et autres
Publié: (2025)

Challenging Vision-Language Models with Physically Deployable Multimodal Semantic Lighting Attacks
par: Zhao, Yingying, et autres
Publié: (2026)

When Surfaces Lie: Exploiting Wrinkle-Induced Attention Shift to Attack Vision-Language Models
par: Hu, Chengyin, et autres
Publié: (2026)

CKDA: Cross-modality Knowledge Disentanglement and Alignment for Visible-Infrared Lifelong Person Re-identification
par: Cui, Zhenyu, et autres
Publié: (2025)

Multi-View Black-Box Physical Attacks on Infrared Pedestrian Detectors Using Adversarial Infrared Grid
par: Tiliwalidi, Kalibinuer, et autres
Publié: (2024)

VisualDAN: Exposing Vulnerabilities in VLMs with Visual-Driven DAN Commands
par: Liu, Aofan, et autres
Publié: (2025)

IV-tuning: Parameter-Efficient Transfer Learning for Infrared-Visible Tasks
par: Zhang, Yaming, et autres
Publié: (2024)

Exposing LLM Vulnerabilities: Adversarial Scam Detection and Performance
par: Chang, Chen-Wei, et autres
Publié: (2024)

Exposing and Addressing Cross-Task Inconsistency in Unified Vision-Language Models
par: Maharana, Adyasha, et autres
Publié: (2023)

Spectral Enhancement and Pseudo-Anchor Guidance for Infrared-Visible Person Re-Identification
par: Ge, Yiyuan, et autres
Publié: (2024)

A Geometric Framework for Adversarial Vulnerability in Machine Learning
par: Bell, Brian
Publié: (2024)

UniRGB-IR: A Unified Framework for Visible-Infrared Semantic Tasks via Adapter Tuning
par: Yuan, Maoxun, et autres
Publié: (2024)

ATFusion: An Alternate Cross-Attention Transformer Network for Infrared and Visible Image Fusion
par: Yan, Han, et autres
Publié: (2024)

Exposing Hallucinations To Suppress Them: VLMs Representation Editing With Generative Anchors
par: Shi, Youxu, et autres
Publié: (2025)

Unified Restoration-Perception Learning: Maritime Infrared-Visible Image Fusion and Segmentation
par: Cai, Weichao, et autres
Publié: (2026)

Agentic Adversarial Rewriting Exposes Architectural Vulnerabilities in Black-Box NLP Pipelines
par: Bethany, Mazal, et autres
Publié: (2026)

Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption
par: Liu, Jinyuan, et autres
Publié: (2025)

IVGF: The Fusion-Guided Infrared and Visible General Framework
par: Liu, Fangcen, et autres
Publié: (2024)

Audio Jailbreak Attacks: Exposing Vulnerabilities in SpeechGPT in a White-Box Framework
par: Ma, Binhao, et autres
Publié: (2025)

UMCFuse: A Unified Multiple Complex Scenes Infrared and Visible Image Fusion Framework
par: Li, Xilai, et autres
Publié: (2024)

UNIV: Unified Foundation Model for Infrared and Visible Modalities
par: Mao, Fangyuan, et autres
Publié: (2025)

Do LLMs and VLMs Share Neurons for Inference? Evidence and Mechanisms of Cross-Modal Transfer
par: Cui, Chenhang, et autres
Publié: (2026)

SimpleFusion: A Simple Fusion Framework for Infrared and Visible Images
par: Chen, Ming, et autres
Publié: (2024)

Geometrically Aligned Transfer Encoder for Inductive Transfer in Regression Tasks
par: Ko, Sung Moon, et autres
Publié: (2023)

Exposing the Cracks: Vulnerabilities of Retrieval-Augmented LLM-based Machine Translation
par: Sun, Yanming, et autres
Publié: (2025)

On the Adversarial Vulnerabilities of Transfer Learning in Remote Sensing
par: Bai, Tao, et autres
Publié: (2025)

DCEvo: Discriminative Cross-Dimensional Evolutionary Learning for Infrared and Visible Image Fusion
par: Liu, Jinyuan, et autres
Publié: (2025)

Exposing Citation Vulnerabilities in Generative Engines
par: Mochizuki, Riku, et autres
Publié: (2025)

Robust SAM: On the Adversarial Robustness of Vision Foundation Models
par: Long, Jiahuan, et autres
Publié: (2025)

Jailbreaking LLMs & VLMs: Mechanisms, Evaluation, and Unified Defense
par: Chen, Zejian, et autres
Publié: (2026)

Isotonic Layer: A Unified Framework for Recommendation Calibration and Debiasing
par: Cheng, Hailing, et autres
Publié: (2026)

Your Vision-Language Model Can't Even Count to 20: Exposing the Failures of VLMs in Compositional Counting
par: Guo, Xuyang, et autres
Publié: (2025)

Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs
par: Hu, Jiayu, et autres
Publié: (2025)

Counterfactual Intervention Feature Transfer for Visible-Infrared Person Re-identification
par: Li, Xulin, et autres
Publié: (2022)