Research Themes

Our research explores the geometry of data and models across multiple areas.

Graph Neural Networks

Understanding and improving GNNs through geometric perspectives, addressing over-smoothing, and developing novel architectures.

Geometric Deep Learning

Leveraging hyperbolic and Riemannian geometry for better representation learning of hierarchical and structured data.

Trustworthy AI

Developing methods for machine unlearning, privacy-preserving ML, and robust evaluation of AI systems.

ML for Science

Applying machine learning to scientific discovery, including drug design, molecular modeling, and computational biology.

Language Models

Research on LLM personalization, retrieval-augmented generation, and geometric reasoning in vision-language models.

Generative Models

Advancing graph generation, diffusion models, and generative approaches for structured data.