Michael Bronstein

papers

281

total citations

papers (10)

SE(3)-Stochastic Flow Matching for Protein Backbone Generation

ICLR 2024arXiv

147

citations

Why do LLMs attend to the first token?

COLM 2025arXiv

citations

Steering Masked Discrete Diffusion Models via Discrete Denoising Posterior Prediction

ICLR 2025arXiv

citations

Multi-domain Distribution Learning for De Novo Drug Design

ICLR 2025arXiv

citations

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

ICLR 2025arXiv

citations

On Measuring Long-Range Interactions in Graph Neural Networks

ICML 2025arXiv

citations

Balancing Efficiency and Expressiveness: Subgraph GNNs with Walk-Based Centrality

ICML 2025arXiv

citations

Michael Bronstein

papers (10)

SE(3)-Stochastic Flow Matching for Protein Backbone Generation

Why do LLMs attend to the first token?

Steering Masked Discrete Diffusion Models via Discrete Denoising Posterior Prediction

Multi-domain Distribution Learning for De Novo Drug Design

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

On Measuring Long-Range Interactions in Graph Neural Networks

Balancing Efficiency and Expressiveness: Subgraph GNNs with Walk-Based Centrality

GradMetaNet: An Equivariant Architecture for Learning on Gradients

Over-squashing in Spatiotemporal Graph Neural Networks

A General Graph Spectral Wavelet Convolution via Chebyshev Order Decomposition

papers (10)

SE(3)-Stochastic Flow Matching for Protein Backbone Generation

Why do LLMs attend to the first token?

Steering Masked Discrete Diffusion Models via Discrete Denoising Posterior Prediction

Multi-domain Distribution Learning for De Novo Drug Design

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

On Measuring Long-Range Interactions in Graph Neural Networks

Balancing Efficiency and Expressiveness: Subgraph GNNs with Walk-Based Centrality

GradMetaNet: An Equivariant Architecture for Learning on Gradients

Over-squashing in Spatiotemporal Graph Neural Networks

A General Graph Spectral Wavelet Convolution via Chebyshev Order Decomposition