5 papers
On the Reliability of Vision-Language Models Under Adversarial Frequency-Domain Perturbations
Jordan Vice, Naveed Akhtar, Yansong Gao +2
Vision-Language Models (VLMs) are increasingly used as perceptual modules for visual content reasoning, including through captioning and DeepFake detection. In this work, we expose…
On the Fairness, Diversity and Reliability of Text-to-Image Generative Models
Jordan Vice, Naveed Akhtar, Leonid Sigal +2
The rapid proliferation of multimodal generative models has sparked critical discussions on their reliability, fairness and potential for misuse. While text-to-image models excel a…
Exploring Bias in over 100 Text-to-Image Generative Models
Jordan Vice, Naveed Akhtar, Richard Hartley +1
We investigate bias trends in text-to-image generative models over time, focusing on the increasing availability of models through open platforms like Hugging Face. While these pla…
Safety Without Semantic Disruptions: Editing-free Safe Image Generation via Context-preserving Dual Latent Reconstruction
Jordan Vice, Naveed Akhtar, Mubarak Shah +2
Training multimodal generative models on large, uncurated datasets can result in users being exposed to harmful, unsafe and controversial or culturally-inappropriate outputs. While…
Manipulating and Mitigating Generative Model Biases without Retraining
Jordan Vice, Naveed Akhtar, Richard Hartley +1
Text-to-image (T2I) generative models have gained increased popularity in the public domain. While boasting impressive user-guided generative abilities, their black-box nature expo…