We can see her presenting on simulations and learning tasks as graphs, with an emphasis on two models: message passing graph neural networks, and graph transformers.

Each message-passing step covers immediate neighbors, and additional layers capture a wider network context, enabling a comprehensive understanding of graph structures. TF-GNN overview Fig. 2: Layers ...

The team proposed a Graph-Segmenter, including a Graph Transformer and a Boundary-aware Attention module, which is an effective network for simultaneously modeling the more profound relation ...