Schema-Agnostic Graph Reasoning Agent for Hybrid Knowledge Graphs

2026-08-16

AI Summary

This study addresses the reasoning challenges arising from unknown schemas in hybrid knowledge graphs by proposing GRA, an intelligent agent employing a schema-agnostic graph reasoning paradigm. Rather than relying on exhaustive context input, GRA dynamically explores the graph through tool invocation and selective navigation mechanisms to enable structured access. Evaluated on the UFK-M benchmark, GRA achieves an accuracy of 88.4%, outperforming full-context methods by 5.1 percentage points while reducing token consumption by over two-thirds. These results demonstrate that selective agentic access significantly surpasses exhaustive processing in resource-constrained scenarios, establishing a novel paradigm for efficient graph reasoning.

Abstract

Tool-calling LLM agents navigate unfamiliar codebases with a handful of generic primitives for listing, reading and searching files (ls, cat, grep). A knowledge graph admits the same interface: listing neighbours, reading node content and searching descriptions are the same operations on a different substrate. Building on this correspondence, we present GRA, a Graph Reasoning Agent that explores hybrid knowledge graphs, whose nodes are either textual concepts or relational tables, with seven generic tools, discovering everything domain-specific at run time. On UFK-M (Unified Factory Knowledge Model), an industrial benchmark of 258 analytical questions whose gold answers are produced by executing validated SQL programs, GRA beats a full-context agent by 5.1 pp (88.4% vs. 83.3%), while reading under a third of its input tokens. A graph-free control shows the gain comes chiefly from selective agentic access rather than graph topology, and that the effect depends on a model able to drive tools reliably. Seeing less, the agent answers better: selective navigation over a structured substrate beats exhaustive context.

Problem

• Hybrid Knowledge Graphs
• Schema-Agnostic Reasoning
• Graph Reasoning Agent
• Tool-calling LLM

Innovation

• Schema-Agnostic
• Graph Reasoning Agent
• Hybrid Knowledge Graphs
• Tool-calling LLM
• Selective Navigation