Do Neural Scaling Laws Exist on Graph Self-Supervised Learning?

Q Ma, H Mao, J Liu, Z Zhang, C Feng, Y Song… - ar** and training scalable graph foundation models (GFM)
using HydraGNN, a multi-headed graph convolutional neural network architecture …

How Local isLocal'? Deep Learning Reveals Locality of the Induced Magnetic Field of Polycyclic Aromatic Hydrocarbons

Y Davidson, A Philipp, S Chakraborty, AM Bronstein… - 2025 - chemrxiv.org
We investigate the locality of magnetic response in polycyclic aromatic molecules using a
novel deep-learning approach. Our method employs graph neural networks (GNNs) with a …

Towards Neural Scaling Laws on Graphs

J Liu, H Mao, Z Chen, T Zhao, N Shah… - The Third Learning on … - openreview.net
Deep graph models (eg, graph neural networks and graph transformers) have become
important techniques for leveraging knowledge across various types of graphs. Yet, the …

Fully-inductive Node Classification on Arbitrary Graphs

J Zhao, M Galkin, H Mostafa, MM Bronstein… - … Models: Evolving AI for … - openreview.net
One fundamental challenge in graph machine learning is generalizing to new graphs. Many
existing methods following the inductive setup can generalize to test graphs with new …