← Search

Dominik Schnaus

2 accepted papers

2025

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

CVPR 2025poster

The platonic representation hypothesis suggests that vision and language embeddings become more homogeneous as model and dataset sizes increase. In particular, pairwise distances within each modality become more similar. This suggests that as foundation models mature, it may become possible to match…

2023

Learning Expressive Priors for Generalization and Uncertainty Estimation in Neural Networks

ICML 2023poster

In this work, we propose a novel prior learning method for advancing generalization and uncertainty estimation in deep neural networks. The key idea is to exploit scalable and structured posteriors of neural networks as informative priors with generalization guarantees. Our learned priors provide ex…