AI RESEARCH
On the Safety of Graph Representation Learning
arXiv CS.LG
•
ArXi:2605.06576v1 Announce Type: new Graph representation learning (GRL) has evolved from topology-only graph embeddings to task-specific supervised GNNs, and recently to reusable representations and graph foundation models (GFMs). However, existing evaluations mainly measure clean transfer, adaptation, and task coverage. It remains unclear whether GRL methods stay reliable when deployment stresses affect graph signals, graph contexts, label, structural groups, or predictive evidence. We