Simone Scardapane

Universal Properties of Activation Sparsity in Modern Large Language Models

MASS: MoErging through Adaptive Subspace Selection

From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Module

Hypergraph Neural Networks through the Lens of Message Passing: A Common Perspective to Homophily and Architecture Design

Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion