Copertina

Visual Memory Injection Attacks for Multi-Turn Conversations

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Schlarmann, Christian, Hein, Matthias
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computer Vision and Pattern Recognition Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

FuseLIP: Multimodal Embeddings via Early Fusion of Discrete Tokens
di: Schlarmann, Christian, et al.
Pubblicazione: (2025)

Adversarially Robust CLIP Models Can Induce Better (Robust) Perceptual Metrics
di: Croce, Francesco, et al.
Pubblicazione: (2025)

Robust CLIP: Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models
di: Schlarmann, Christian, et al.
Pubblicazione: (2024)

Robustness in Both Domains: CLIP Needs a Robust Text Encoder
di: Rocamora, Elias Abad, et al.
Pubblicazione: (2025)

On the Out-of-Distribution Generalization of Reasoning in Multimodal LLMs for Simple Visual Planning Tasks
di: Neuhaus, Yannic, et al.
Pubblicazione: (2026)

Mahalanobis++: Improving OOD Detection via Feature Normalization
di: Mueller, Maximilian, et al.
Pubblicazione: (2025)

How to train your ViT for OOD Detection
di: Mueller, Maximilian, et al.
Pubblicazione: (2024)

LoGex: Improved tail detection of extremely rare histopathology classes via guided diffusion
di: Mueller, Maximilian, et al.
Pubblicazione: (2024)

DiG-IN: Diffusion Guidance for Investigating Networks -- Uncovering Classifier Differences Neuron Visualisations and Visual Counterfactual Explanations
di: Augustin, Maximilian, et al.
Pubblicazione: (2023)

Memory-V2V: Memory-Augmented Video-to-Video Diffusion for Consistent Multi-Turn Editing
di: Lee, Dohun, et al.
Pubblicazione: (2026)

Advancing Compositional Awareness in CLIP with Efficient Fine-Tuning
di: Peleg, Amit, et al.
Pubblicazione: (2025)

Towards Reliable Evaluation and Fast Training of Robust Semantic Segmentation Models
di: Croce, Francesco, et al.
Pubblicazione: (2023)

RePOPE: Impact of Annotation Errors on the POPE Benchmark
di: Neuhaus, Yannic, et al.
Pubblicazione: (2025)

eXIAA: eXplainable Injections for Adversarial Attack
di: Pesce, Leonardo, et al.
Pubblicazione: (2025)

PVI: Plug-in Visual Injection for Vision-Language-Action Models
di: Zhang, Zezhou, et al.
Pubblicazione: (2026)

Time-, Memory- and Parameter-Efficient Visual Adaptation
di: Mercea, Otniel-Bogdan, et al.
Pubblicazione: (2024)

Mind the Detail: Uncovering Clinically Relevant Image Details in Accelerated MRI with Semantically Diverse Reconstructions
di: Morshuis, Jan Nikolas, et al.
Pubblicazione: (2025)

U-Turn Diffusion
di: Behjoo, Hamidreza, et al.
Pubblicazione: (2023)

Tiny Neural Networks for Multi-Object Tracking in a Modular Kalman Framework
di: Holz, Christian Alexander, et al.
Pubblicazione: (2025)

Learning from Memory: Non-Parametric Memory Augmented Self-Supervised Learning of Visual Features
di: Silva, Thalles, et al.
Pubblicazione: (2024)

Toward Inherently Robust VLMs Against Visual Perception Attacks
di: MohajerAnsari, Pedram, et al.
Pubblicazione: (2025)

WebInject: Prompt Injection Attack to Web Agents
di: Wang, Xilong, et al.
Pubblicazione: (2025)

DASH: Detection and Assessment of Systematic Hallucinations of VLMs
di: Augustin, Maximilian, et al.
Pubblicazione: (2025)

Memory Backdoor Attacks on Neural Networks
di: Luzon, Eden, et al.
Pubblicazione: (2024)

Familiarity-Based Open-Set Recognition Under Adversarial Attacks
di: Enevoldsen, Philip, et al.
Pubblicazione: (2023)

WebLINX: Real-World Website Navigation with Multi-Turn Dialogue
di: Lù, Xing Han, et al.
Pubblicazione: (2024)

Perturb and Recover: Fine-tuning for Effective Backdoor Removal from CLIP
di: Singh, Naman Deep, et al.
Pubblicazione: (2024)

Beyond Knowledge Silos: Task Fingerprinting for Democratization of Medical Imaging AI
di: Godau, Patrick, et al.
Pubblicazione: (2024)

LanteRn: Latent Visual Structured Reasoning
di: Viveiros, André G., et al.
Pubblicazione: (2026)

Tuning Just Enough: Lightweight Backdoor Attacks on Multi-Encoder Diffusion Models
di: Chen, Ziyuan, et al.
Pubblicazione: (2026)

Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification
di: Sun, Yuhao, et al.
Pubblicazione: (2025)

External Knowledge Injection for CLIP-Based Class-Incremental Learning
di: Zhou, Da-Wei, et al.
Pubblicazione: (2025)

Proactive Agents for Multi-Turn Text-to-Image Generation Under Uncertainty
di: Hahn, Meera, et al.
Pubblicazione: (2024)

SURE-VQA: Systematic Understanding of Robustness Evaluation in Medical VQA Tasks
di: Kahl, Kim-Celine, et al.
Pubblicazione: (2024)

Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuning
di: Jie, Shibo, et al.
Pubblicazione: (2024)

Infinite dSprites for Disentangled Continual Learning: Separating Memory Edits from Generalization
di: Dziadzio, Sebastian, et al.
Pubblicazione: (2023)

Multi-Task Learning for Visually Grounded Reasoning in Gastrointestinal VQA
di: Safwan, Itbaan, et al.
Pubblicazione: (2025)

MMDU: A Multi-Turn Multi-Image Dialog Understanding Benchmark and Instruction-Tuning Dataset for LVLMs
di: Liu, Ziyu, et al.
Pubblicazione: (2024)

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

Memory Efficient Full-gradient Attacks (MEFA) Framework for Adversarial Defense Evaluations
di: Du, Yuan, et al.
Pubblicazione: (2026)