Search

Roland S Zimmermann

In search of forgotten domain generalization
InfoNCE: Identifying the Gap Between Theory and Practice
Scale Alone Does not Improve Mechanistic Interpretability in Vision Models
Don't trust your eyes: on the (un) reliability of feature visualizations
Provably Learning Object-Centric Representations
Increasing Confidence in Adversarial Robustness Evaluations
Content suppresses style: dimensionality collapse in contrastive learning
How Well do Feature Visualizations Support Causal Understanding of CNN Activations?
Contrastive Learning Inverts the Data Generating Process
Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations
A simple way to make neural networks robust against diverse image corruptions

Published with Wowchemy — the free, open source website builder that empowers creators.