Ali Jadbabaie

Is Your Diffusion Model Actually Denoising?

Residual Connections and Normalization Can Provably Prevent Oversmoothing in GNNs

On the Emergence of Position Bias in Transformers

Fast Tensor Completion via Approximate Richardson Iteration

Linear attention is (maybe) all you need (to understand Transformer optimization)

On the Role of Attention Masks and LayerNorm in Transformers