Imagen de Portada

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

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Chen, Xiang, Dong, Yuxian, Li, Chao, Hu, Chengyin, Han, Jiaju, Zhang, Fengyu, Wei, Yiwei, Long, Jiahuan, Guo, Jiujiang
Formato:	Preprint
Publicado:	2026
Materias:	Computer Vision and Pattern Recognition
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

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

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

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

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

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

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

CDUPatch: Color-Driven Universal Adversarial Patch Attack for Dual-Modal Visible-Infrared Detectors
por: Long, Jiahuan, et al.
Publicado: (2025)

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

When Surfaces Lie: Exploiting Wrinkle-Induced Attention Shift to Attack Vision-Language Models
por: Hu, Chengyin, et al.
Publicado: (2026)

CKDA: Cross-modality Knowledge Disentanglement and Alignment for Visible-Infrared Lifelong Person Re-identification
por: Cui, Zhenyu, et al.
Publicado: (2025)

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

VisualDAN: Exposing Vulnerabilities in VLMs with Visual-Driven DAN Commands
por: Liu, Aofan, et al.
Publicado: (2025)

IV-tuning: Parameter-Efficient Transfer Learning for Infrared-Visible Tasks
por: Zhang, Yaming, et al.
Publicado: (2024)

Exposing LLM Vulnerabilities: Adversarial Scam Detection and Performance
por: Chang, Chen-Wei, et al.
Publicado: (2024)

Exposing and Addressing Cross-Task Inconsistency in Unified Vision-Language Models
por: Maharana, Adyasha, et al.
Publicado: (2023)

Spectral Enhancement and Pseudo-Anchor Guidance for Infrared-Visible Person Re-Identification
por: Ge, Yiyuan, et al.
Publicado: (2024)

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

UniRGB-IR: A Unified Framework for Visible-Infrared Semantic Tasks via Adapter Tuning
por: Yuan, Maoxun, et al.
Publicado: (2024)

ATFusion: An Alternate Cross-Attention Transformer Network for Infrared and Visible Image Fusion
por: Yan, Han, et al.
Publicado: (2024)

Exposing Hallucinations To Suppress Them: VLMs Representation Editing With Generative Anchors
por: Shi, Youxu, et al.
Publicado: (2025)

Unified Restoration-Perception Learning: Maritime Infrared-Visible Image Fusion and Segmentation
por: Cai, Weichao, et al.
Publicado: (2026)

Agentic Adversarial Rewriting Exposes Architectural Vulnerabilities in Black-Box NLP Pipelines
por: Bethany, Mazal, et al.
Publicado: (2026)

Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption
por: Liu, Jinyuan, et al.
Publicado: (2025)

IVGF: The Fusion-Guided Infrared and Visible General Framework
por: Liu, Fangcen, et al.
Publicado: (2024)

Audio Jailbreak Attacks: Exposing Vulnerabilities in SpeechGPT in a White-Box Framework
por: Ma, Binhao, et al.
Publicado: (2025)

UMCFuse: A Unified Multiple Complex Scenes Infrared and Visible Image Fusion Framework
por: Li, Xilai, et al.
Publicado: (2024)

UNIV: Unified Foundation Model for Infrared and Visible Modalities
por: Mao, Fangyuan, et al.
Publicado: (2025)

Do LLMs and VLMs Share Neurons for Inference? Evidence and Mechanisms of Cross-Modal Transfer
por: Cui, Chenhang, et al.
Publicado: (2026)

SimpleFusion: A Simple Fusion Framework for Infrared and Visible Images
por: Chen, Ming, et al.
Publicado: (2024)

Geometrically Aligned Transfer Encoder for Inductive Transfer in Regression Tasks
por: Ko, Sung Moon, et al.
Publicado: (2023)

Exposing the Cracks: Vulnerabilities of Retrieval-Augmented LLM-based Machine Translation
por: Sun, Yanming, et al.
Publicado: (2025)

On the Adversarial Vulnerabilities of Transfer Learning in Remote Sensing
por: Bai, Tao, et al.
Publicado: (2025)

DCEvo: Discriminative Cross-Dimensional Evolutionary Learning for Infrared and Visible Image Fusion
por: Liu, Jinyuan, et al.
Publicado: (2025)

Exposing Citation Vulnerabilities in Generative Engines
por: Mochizuki, Riku, et al.
Publicado: (2025)

Jailbreaking LLMs & VLMs: Mechanisms, Evaluation, and Unified Defense
por: Chen, Zejian, et al.
Publicado: (2026)

Robust SAM: On the Adversarial Robustness of Vision Foundation Models
por: Long, Jiahuan, et al.
Publicado: (2025)

Isotonic Layer: A Unified Framework for Recommendation Calibration and Debiasing
por: Cheng, Hailing, et al.
Publicado: (2026)

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

Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs
por: Hu, Jiayu, et al.
Publicado: (2025)

Counterfactual Intervention Feature Transfer for Visible-Infrared Person Re-identification
por: Li, Xulin, et al.
Publicado: (2022)