11 citations · 47 across the 48 of their papers we have counts for
4 papers · 2 filters
Dyslexify: A Mechanistic Defense Against Typographic Attacks in CLIP
Lorenz Hufe, Constantin Venhoff, Erblina Purelku +3
Typographic attacks exploit multi-modal systems by injecting text into images, leading to targeted misclassifications, malicious content generation and even Vision-Language Model j…
Deep Learning-based Multi Project InP Wafer Simulation for Unsupervised Surface Defect Detection
Emílio Dolgener Cantú, Rolf Klemens Wittmann, Oliver Abdeen +4
Quality management in semiconductor manufacturing often relies on template matching with known golden standards. For Indium-Phosphide (InP) multi-project wafer manufacturing, low p…
Steering CLIP's vision transformer with sparse autoencoders
Sonia Joseph, Praneet Suresh, Ethan Goldfarb +6
While vision models are highly capable, their internal mechanisms remain poorly understood -- a challenge which sparse autoencoders (SAEs) have helped address in language, but whic…
Post-Hoc Concept Disentanglement: From Correlated to Isolated Concept Representations
Eren Erogullari, Sebastian Lapuschkin, Wojciech Samek +1
Concept Activation Vectors (CAVs) are widely used to model human-understandable concepts as directions within the latent space of neural networks. They are trained by identifying d…