GraphOracle: Efficient Fully-Inductive Knowledge Graph Reasoning via Relation-Dependency Graphs

TL;DR

GraphOracle introduces a relation-dependency graph approach for fully-inductive knowledge graph reasoning, enabling accurate predictions on unseen entities and relations with improved efficiency and performance across numerous benchmarks.

Contribution

The paper presents a novel relation-dependency graph framework that enhances fully-inductive reasoning in knowledge graphs by reducing graph complexity and leveraging attention mechanisms.

Findings

Outperforms prior methods by up to 25% in fully-inductive settings.

Achieves up to 28% improvement in cross-domain scenarios.

Efficiently generalizes to unseen entities and relations.

Abstract

Knowledge graph reasoning in the fully-inductive setting, where both entities and relations at test time are unseen during training, remains an open challenge. In this work, we introduce GraphOracle, a novel framework that achieves robust fully-inductive reasoning by transforming each knowledge graph into a Relation-Dependency Graph (RDG). The RDG encodes directed precedence links between relations, capturing essential compositional patterns while drastically reducing graph density. Conditioned on a query relation, a multi-head attention mechanism propagates information over the RDG to produce context-aware relation embeddings. These embeddings then guide a second GNN to perform inductive message passing over the original knowledge graph, enabling prediction on entirely new entities and relations. Comprehensive experiments on 60 benchmarks demonstrate that GraphOracle outperforms prior methods by up to 25% in fully-inductive and 28% in cross-domain scenarios. Our analysis further confirms that the compact RDG structure and attention-based propagation are key to efficient and accurate generalization.