Projects

Symmetry Profiles of Vision Foundation Models: What Makes Visual Representations Useful?

Nikita Araslanov

An alternative perspective on Information Propagation in Graph Neural Networks

Giuseppe Alessio D’Inverno

Text Guided Generative Models for Conditional 3D Material Generation

Exchangeability-Guided Graph Condensation

Graph Curvature as a Lens on Adversarial Robustness in GNNs

Sofiane Ennadir, Oleg Smirnov

AnisoSphere: Learning Anisotropic Geodesic Convolutions for Global Weather Prediction

Simone Foti, Massimiliano Esposito

Equivariant networks alias too!?

Julia Grabinski

Risk-Sensitive Policy Learning over Parametric Linear Operators

Learning Physics on Meshes: Geometry-Aware Neural Operators

Hidden in Plain Symmetry: Exploiting Molecular Symmetry for 3D Canonicalization

New point-cloud architectures for mesh-invariant operator learning

Posterior Geometry and Variational Inference for Bayesian Complexity in Neural Networks

Jiayi Li, Martin Trapp

Neural Generative Models for Temporal Graphs

Hypergraph Heat Method

Jiří Minarčík

Singular learning theory for LLM interpretability and alignment

Sergio Estan Ruiz

Spectral Hypergraph Learning for Higher-order Biological Interactions

Khaled Mohammed Saifuddin

Hierarchy-Aware Training for Generative Retrieval

Topology-Aware Learning for Spatial Transcriptomics

Graph Readouts with Laplacian Projection Valued Measures

Ka Man (Ambrose) Yim