Generalizable Insights for Graph Transformers in Theory and Practice

NeurIPSSpotlight2025

Authors
Timo Stoll, Luis Müller, Christopher Morris
Venue
NeurIPS 2025
Track
Spotlight

TL;DR

Graph Transformers (GTs) have shown strong empirical performance, yet current architectures vary widely in their use of attention mechanisms, positional embeddings (PEs), and expressivity…

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

transformer attention theory graph

Related NeurIPS papers

← All NeurIPS 2025 Spotlight papers · Browse the whole archive