Pierre Vandergheynst

Carré du champ flow matching: better quality-generalisation tradeoff in generative models

Beyond Ensembles: Simulating All-Atom Protein Dynamics in a Learned Latent Space

Return of ChebNet: Understanding and Improving an Overlooked GNN on Long Range Tasks

On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learning

Boosting Protein Graph Representations through Static-Dynamic Fusion

Generative Modeling of Full-Atom Protein Conformations using Latent Diffusion on Graph Embeddings