Publications

On the Emergence of Position Bias in Transformers

What Are We Optimizing For? A Human-centric Evaluation of Deep Learning-based Movie Recommenders

Residual Connections and Normalization Can Provably Prevent Oversmoothing in GNNs

GraphHash: Graph Clustering Enables Parameter Efficiency in Recommender Systems

Understanding and Scaling Collaborative Filtering Optimization from the Perspective of Matrix Rank

On the Role of Attention Masks and LayerNorm in Transformers

Demystifying Oversmoothing in Attention-Based Graph Neural Networks

An Non-Asymptotic Analysis of Oversmoothing in Graph Neural Networks

Link Partitioning on Simplicial Complexes Using Higher-Order Laplacians

Spectra of Quantum Graphs via a Scattering Matrix Approach