Yoav Gelberg

On The Expressive Power of GNN Derivatives

Extending the Context of Pretrained LLMs by Dropping Their Positional Embedding

Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity

GradMetaNet: An Equivariant Architecture for Learning on Gradients

Learning on LoRAs: GL-Equivariant Processing of Low-Rank Weight Spaces for Large Finetuned Models