arXiv:2502. 16533v3 Announce Type: replace-cross Abstract: Graph Transformers (GTs) have demonstrated a strong capability in modeling graph structures by addressing the intrinsic limitations of graph neural networks (GNNs), such as over-smoothing and over-squashing.
The paper explores pre‑training methods for graph transformers applied to biochemistry. It finds that supervised pre‑training with computed property labels yields the best improvement on downstream tasks, and it emphasizes that limiting model capacity helps prevent overfitting.
The paper shows that a plain Transformer can serve as an effective graph learner by adding three lightweight modifications: simplified L₂ attention, adaptive RMS normalization, and an MLP-based positional encoding stem. These changes preserve token magnitude and enable the model to achieve high expressivity on graph benchmarks, outperforming more complex graph transformer variants. The results suggest that plain Transformers can act as a unified backbone for multimodal learning across language, vision, and graph domains.
arXiv:2412. 19419v2 Announce Type: replace-cross Abstract: Graph neural networks are deep neural networks designed for graphs with attributes attached to nodes or edges.
arXiv:2607. 03109v1 Announce Type: cross Abstract: We study a graph classification problem involving over 20 million graphs, arising from high-order perturbative computations of correlators in planar $\mathcal{N}=4$ super-Yang--Mills, a model closely related to the theory of the strong nuclear force.
arXiv:2509. 22259v4 Announce Type: replace-cross Abstract: We study the extent to which rotary position encodings (RoPE), a recent transformer position encoding algorithm broadly adopted in large language models (LLMs) and vision transformers (ViTs), can be applied to graph-structured data.
arXiv:2602. 01553v3 Announce Type: replace-cross Abstract: Link prediction is a core challenge in graph machine learning, demanding models that capture rich and complex topological dependencies.
arXiv:2602.15239v3 Announce Type: replace Abstract: Transformers have achieved remarkable success across domains, motivating the rise of Graph Transformers (GTs) as attention-based architectures for...
The paper investigates Graph Transformers (GTs) that use Graph Neural Network (GNN)-based positional encodings. By analyzing GTs through manifold limit models, the authors establish a theoretical link between GTs and Manifold Neural Networks (MNNs), showing that GTs inherit transferability guarantees from their positional encodings. Experiments on standard graph benchmarks confirm that GTs scale comparably to GNNs, and a real‑world shortest‑path estimation task demonstrates their practical efficiency on large terrains.
The paper investigates how the choice of graph tokenization affects transformer expressivity. It analyzes three tokenization families—spectral, random‑walk, and adjacency—showing that each induces different depth requirements and that some tokenizations are inherently lossy or ill‑conditioned for certain tasks. The authors prove lower bounds and impossibility results for converting between tokenizations and validate these findings with experiments on synthetic and real‑world data.
The paper introduces a new (pseudo)-metric called graph homomorphism distortion, which quantifies how much node features must be distorted when mapping one attributed graph to another. This metric builds on metric geometry concepts and leverages graph homomorphisms to capture both structure and features. The authors show that the metric can be efficiently approximated, offers fine-grained graph distinction, and can serve as a structural encoding that enhances graph neural network predictions.