A single input-agnostic thermal-airflow perturbation, optimized on one surrogate CLIP model, transfers to five CLIP backbones and six VLMs, degrading scene classification by up to 38.2%.
In: International conference on machine learning
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
A^4D detects adversarial attacks in an attack- and classifier-agnostic way by measuring non-arbitrary shifts in CLIP embedding space from prompt-based similarity scores.
Hybrid quantum-classical models using structured entanglement keep high accuracy on MNIST, OrganAMNIST and CIFAR-10 while lowering adversarial attack success rates and raising the computational cost of generating attacks.
citing papers explorer
-
AirflowAttack: Thermal-Airflow Adversarial Perturbations against Infrared Remote-Sensing Vision-Language Models
A single input-agnostic thermal-airflow perturbation, optimized on one surrogate CLIP model, transfers to five CLIP backbones and six VLMs, degrading scene classification by up to 38.2%.
-
A Classifier-Agnostic Zero-Shot Adversarial Attack Detection via CLIP
A^4D detects adversarial attacks in an attack- and classifier-agnostic way by measuring non-arbitrary shifts in CLIP embedding space from prompt-based similarity scores.
-
QShield: Securing Neural Networks Against Adversarial Attacks using Quantum Circuits
Hybrid quantum-classical models using structured entanglement keep high accuracy on MNIST, OrganAMNIST and CIFAR-10 while lowering adversarial attack success rates and raising the computational cost of generating attacks.