Technical Report: The Graph Spectral Token -- Enhancing Graph Transformers with Spectral Information
This work addresses a significant bottleneck in graph transformers for researchers and practitioners in graph learning, though it is incremental as it builds on existing models like GraphTrans and SubFormer.
The authors tackled the challenge of incorporating graph inductive bias into transformer architectures by proposing the Graph Spectral Token to encode spectral information, resulting in over 10% improvements on large graph benchmarks while maintaining efficiency comparable to MP-GNNs.
Graph Transformers have emerged as a powerful alternative to Message-Passing Graph Neural Networks (MP-GNNs) to address limitations such as over-squashing of information exchange. However, incorporating graph inductive bias into transformer architectures remains a significant challenge. In this report, we propose the Graph Spectral Token, a novel approach to directly encode graph spectral information, which captures the global structure of the graph, into the transformer architecture. By parameterizing the auxiliary [CLS] token and leaving other tokens representing graph nodes, our method seamlessly integrates spectral information into the learning process. We benchmark the effectiveness of our approach by enhancing two existing graph transformers, GraphTrans and SubFormer. The improved GraphTrans, dubbed GraphTrans-Spec, achieves over 10% improvements on large graph benchmark datasets while maintaining efficiency comparable to MP-GNNs. SubFormer-Spec demonstrates strong performance across various datasets.