Imagen de Portada

Modal Aphasia: Can Unified Multimodal Models Describe Images From Memory?

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Aerni, Michael, Swanson, Joshua, Nikolić, Kristina, Tramèr, Florian
Formato:	Preprint
Publicado:	2025
Materias:	Computer Vision and Pattern Recognition Cryptography and Security
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Image Can Bring Your Memory Back: A Novel Multi-Modal Guided Attack against Image Generation Model Unlearning
por: Liu, Renyang, et al.
Publicado: (2025)

Evaluations of Machine Learning Privacy Defenses are Misleading
por: Aerni, Michael, et al.
Publicado: (2024)

Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application
por: Xu, Wenzhuo, et al.
Publicado: (2025)

A Cross-Modal Prompt Injection Attack against Large Vision-Language Models with Image-Only Perturbation
por: Yang, Hao, et al.
Publicado: (2026)

Large-scale online deanonymization with LLMs
por: Lermen, Simon, et al.
Publicado: (2026)

Fusion is Not Enough: Single Modal Attacks on Fusion Models for 3D Object Detection
por: Cheng, Zhiyuan, et al.
Publicado: (2023)

Membership Inference Attacks on Sequence Models
por: Rossi, Lorenzo, et al.
Publicado: (2025)

OmniSafeBench-MM: A Unified Benchmark and Toolbox for Multimodal Jailbreak Attack-Defense Evaluation
por: Jia, Xiaojun, et al.
Publicado: (2025)

Unbridled Icarus: A Survey of the Potential Perils of Image Inputs in Multimodal Large Language Model Security
por: Fan, Yihe, et al.
Publicado: (2024)

MMA-Diffusion: MultiModal Attack on Diffusion Models
por: Yang, Yijun, et al.
Publicado: (2023)

SoK: Can Synthetic Images Replace Real Data? A Survey of Utility and Privacy of Synthetic Image Generation
por: Chung, Yunsung, et al.
Publicado: (2025)

From Evidence to Verdict: An Agent-Based Forensic Framework for AI-Generated Image Detection
por: Liang, Mengfei, et al.
Publicado: (2025)

Solutions to Deepfakes: Can Camera Hardware, Cryptography, and Deep Learning Verify Real Images?
por: Vilesov, Alexander, et al.
Publicado: (2024)

Evolving Contextual Safety in Multi-Modal Large Language Models via Inference-Time Self-Reflective Memory
por: Zhang, Ce, et al.
Publicado: (2026)

Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt
por: Ying, Zonghao, et al.
Publicado: (2024)

FreqCross: A Multi-Modal Frequency-Spatial Fusion Network for Robust Detection of Stable Diffusion 3.5 Generated Images
por: Yang, Guang
Publicado: (2025)

When Memory Becomes a Vulnerability: Towards Multi-turn Jailbreak Attacks against Text-to-Image Generation Systems
por: Zhao, Shiqian, et al.
Publicado: (2025)

Purified and Unified Steganographic Network
por: Li, Guobiao, et al.
Publicado: (2024)

Provably Secure Robust Image Steganography via Cross-Modal Error Correction
por: Qi, Yuang, et al.
Publicado: (2024)

Universally Unfiltered and Unseen:Input-Agnostic Multimodal Jailbreaks against Text-to-Image Model Safeguards
por: Yan, Song, et al.
Publicado: (2025)

VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models
por: Yin, Ziyi, et al.
Publicado: (2023)

AMB-FHE: Adaptive Multi-biometric Fusion with Fully Homomorphic Encryption
por: Bayer, Florian, et al.
Publicado: (2025)

From Easy to Hard++: Promoting Differentially Private Image Synthesis Through Spatial-Frequency Curriculum
por: Gong, Chen, et al.
Publicado: (2026)

Espresso: Robust Concept Filtering in Text-to-Image Models
por: Das, Anudeep, et al.
Publicado: (2024)

Membership Inference Attack Against Masked Image Modeling
por: Li, Zheng, et al.
Publicado: (2024)

Where the Devil Hides: Deepfake Detectors Can No Longer Be Trusted
por: Yuan, Shuaiwei, et al.
Publicado: (2025)

UVL2: A Unified Framework for Video Tampering Localization
por: Pei, Pengfei
Publicado: (2023)

InverTune: Removing Backdoors from Multimodal Contrastive Learning Models via Trigger Inversion and Activation Tuning
por: Sun, Mengyuan, et al.
Publicado: (2025)

On the Generation and Mitigation of Harmful Geometry in Image-to-3D Models
por: Liu, Yule, et al.
Publicado: (2026)

Federated Learning for Large Models in Medical Imaging: A Comprehensive Review
por: Sun, Mengyu, et al.
Publicado: (2025)

Token-Level Constraint Boundary Search for Jailbreaking Text-to-Image Models
por: Liu, Jiangtao, et al.
Publicado: (2025)

Gungnir: Exploiting Stylistic Features in Images for Backdoor Attacks on Diffusion Models
por: Zhang, Lei, et al.
Publicado: (2025)

Patronus: Safeguarding Text-to-Image Models against White-Box Adversaries
por: Li, Xinfeng, et al.
Publicado: (2025)

Neighbor-Aware Localized Concept Erasure in Text-to-Image Diffusion Models
por: Shi, Zhuan, et al.
Publicado: (2026)

Decomposing Private Image Generation via Coarse-to-Fine Wavelet Modeling
por: Bayrooti, Jasmine, et al.
Publicado: (2026)

HTS-Attack: Heuristic Token Search for Jailbreaking Text-to-Image Models
por: Gao, Sensen, et al.
Publicado: (2024)

Gaussian Shading: Provable Performance-Lossless Image Watermarking for Diffusion Models
por: Yang, Zijin, et al.
Publicado: (2024)

Stable Signature is Unstable: Removing Image Watermark from Diffusion Models
por: Hu, Yuepeng, et al.
Publicado: (2024)

FIDAVL: Fake Image Detection and Attribution using Vision-Language Model
por: Keita, Mamadou, et al.
Publicado: (2024)

Intriguing Properties of Diffusion Models: An Empirical Study of the Natural Attack Capability in Text-to-Image Generative Models
por: Sato, Takami, et al.
Publicado: (2023)