Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Bridging Input Feature Spaces Towards Graph Foundation Models

About

Unlike vision and language domains, graph learning lacks a shared input space, as input features differ across graph datasets not only in semantics, but also in value ranges and dimensionality. This misalignment prevents graph models from generalizing across datasets, limiting their use as foundation models. In this work, we propose ALL-IN, a simple and theoretically grounded method that enables transferability across datasets with different input features. Our approach projects node features into a shared random space and constructs representations via covariance-based statistics, thus eliminating dependence on the original feature space. We show that the computed node-covariance operators and the resulting node representations are invariant in distribution to permutations of the input features. We further demonstrate that the expected operator exhibits invariance to general orthogonal transformations of the input features. Empirically, ALL-IN achieves strong performance across diverse node- and graph-level tasks on unseen datasets with new input features, without requiring architecture changes or retraining. These results point to a promising direction for input-agnostic, transferable graph models.

Moshe Eliasof, Krishna Sri Ipsit Mantri, Beatrice Bevilacqua, Bruno Ribeiro, Carola-Bibiane Sch\"onlieb• 2026

Related benchmarks

TaskDatasetResultRank
Graph ClassificationPROTEINS
Accuracy78.2
1252
Graph ClassificationMUTAG
Accuracy92.9
1103
Node ClassificationChameleon
Accuracy67.4
867
Node ClassificationSquirrel
Accuracy49.98
786
Node ClassificationPubmed
Accuracy78.03
627
Node ClassificationCora
Accuracy82.13
583
Node ClassificationActor
Accuracy29.47
556
Node Classificationogbn-arxiv (test)
Accuracy75.27
497
Node Classificationamazon-ratings
Accuracy49.02
309
Graph ClassificationIMDB-B
Mean Accuracy77.2
159
Showing 10 of 23 rows

Other info

Follow for update