Graphlets as Building Blocks for Structural Vocabulary in Knowledge Graph Foundation Models

TL;DR

This paper introduces a graphlet-based structural vocabulary for Knowledge Graph Foundation Models, enabling better transferability and improved link prediction across diverse KGs.

Contribution

It proposes a universal graphlet vocabulary to capture structural invariances, enhancing transferability of KGFMs across unseen knowledge graphs.

Findings

Graphlet vocabulary improves zero-shot link prediction performance.

Models with graphlet vocabulary outperform prior KGFMs.

Framework is evaluated on 51 diverse knowledge graphs.

Abstract

Foundation models excel at language, where sentences become tokens, and vision, where images become pixels, because both reduce to discrete symbols on a shared, fixed grid. Knowledge Graphs share the discreteness, but not the geometry. Their entities and relations are discrete symbols, yet their arrangement is relational and lacks a common, fixed grid. Knowledge Graphs (KGs) share the discreteness, but not the geometry. They form irregular, non-Euclidean topologies whose local neighborhoods differ from graph to graph. Therefore, Knowledge Graph Foundation Models (KGFMs) rely on identifying structural invariances to produce transferable representations. Without a universal token set, KGFMs are limited in their ability to transfer representations across unseen KGs. We close this gap by treating graphlets, small connected graphs, as structural tokens that recur in heterogeneous KGs. In this paper, We introduce a model-agnostic framework based on a vocabulary of graphlets that mines a KG between relations via pattern matching. In particular, we considered closed and open 2- and 3-path, and star graphlets, to obtain robust invariances. The framework is evaluated on 51 KGs from a wide range of domains, for zero-shot inductive and transductive link prediction. Experiments show that adding simple graphlets to the vocabulary yields models that outperform prior KGFMs.