GraphScholar: Compositional Strategic Decision Making for Knowledge Graph Reasoning via Relation-Dependency Graphs

Zhonglin Jiang, Hange Zhou, Enjun Du, Xuefeng, Sihao Chen, yingjie cui, Jingcheng Sha, Yong Chen, Yongqi Zhang

Keywords: AI for Mechanism Design; Strategic Interaction Modeling; Knowledge Graph Reasoning; Inductive and Cross-Domain Generalization; Compositional Structures; Scalable AI Systems

TL;DR: GRAPHSCHOLAR reframes knowledge graph reasoning as a compositional strategic decision process, using relation-dependency graphs to generalize across unseen entities, relations, and domains.

Abstract: Strategic decision making in complex, dynamic multi-agent environments often depends on reliable relational reasoning under distribution shift, where both the underlying entities and the interaction types can change over time or across domains. 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 \textsc{GraphScholar}, 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 \textsc{GraphScholar} 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.

Track: Long Paper

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 6