English

On the Safety of Graph Representation Learning

Machine Learning 2026-05-08 v1

Abstract

Graph representation learning (GRL) has evolved from topology-only graph embeddings to task-specific supervised GNNs, and more 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 support, structural groups, or predictive evidence. We introduce GRL-Safety, a multi-axis safety evaluation benchmark for GRL. GRL-Safety evaluates twelve representative methods, spanning topology-only embedding methods, supervised GNNs, self-supervised graph models, and GFMs, on twenty-five graph datasets under standardized evaluation conditions while preserving method-native adaptation. The evaluation covers five safety axes: corruption robustness, OOD generalization, class imbalance, fairness, and interpretation, with per-axis and sub-condition reporting rather than a single aggregate score. Our analysis yields three cross-axis insights that can inspire future research. First, safety behavior is shaped by the interaction between representation design and the stressed graph factor, rather than by method family alone. Second, foundation-era methods show axis-specific strengths rather than broad safety dominance. Third, several deployment regimes remain difficult even for the best evaluated method, revealing capability gaps that require new robustness, adaptation, or training objectives beyond model selection. The benchmark, evaluation protocols, and code are available at: https://github.com/GXG-CS/GRL-Safety.

Keywords

Cite

@article{arxiv.2605.06576,
  title  = {On the Safety of Graph Representation Learning},
  author = {Xiaoguang Guo and Zehong Wang and Ziming Li and Shawn Spitzel and Soonwoo Kwon and Tianyi Ma and Yanfang Ye and Chuxu Zhang},
  journal= {arXiv preprint arXiv:2605.06576},
  year   = {2026}
}

Comments

Preprint. 10 pages main text, appendices included

R2 v1 2026-07-01T12:55:37.470Z