Hypergraph Enterprise Agentic Reasoner over Heterogeneous Business Systems

2026-05-13

AI Summary

This work addresses the challenges of semantic grounding and auditability in large language models (LLMs) when deployed in heterogeneous enterprise systems, where hallucinations and failures in multi-hop n-ary reasoning are prevalent. To overcome these limitations, the authors propose a hierarchical hypergraph ontology framework that integrates layer virtualization for interfacing with diverse data sources and procedural hyperedges to encode business rules. Coupled with an evidence-driven dynamic reasoning loop, this approach enables structured multi-hop analysis without requiring LLM retraining. Evaluated on complex enterprise tasks such as root cause analysis of supply chain order fulfillment bottlenecks, the method achieves 94.7% accuracy, substantially reduces token consumption, and automates manual diagnostic workflows while maintaining high interpretability and auditability.

Abstract

Applying Large Language Models (LLMs) to heterogeneous enterprise systems is hindered by hallucinations and failures in multi-hop, n-ary reasoning. Existing paradigms (e.g., GraphRAG, NL2SQL) lack the semantic grounding and auditable execution required for these complex environments. We introduce HEAR, an enterprise agentic reasoner built on a Stratified Hypergraph Ontology. Its base Graph Layer virtualizes provenance-aware data interfaces, while the Hyperedge Layer encodes n-ary business rules and procedural protocols. Operating an evidence-driven reasoning loop, HEAR dynamically orchestrates ontology tools for structured multi-hop analysis without requiring LLM retraining. Evaluations on supply-chain tasks, including order fulfillment blockage root cause analysis (RCA), show HEAR achieves up to 94.7% accuracy. Crucially, HEAR demonstrates adaptive efficiency: utilizing procedural hyperedges to minimize token costs, while leveraging topological exploration for rigorous correctness on complex queries. By matching proprietary model performance with open-weight backbones and automating manual diagnostics, HEAR establishes a scalable, auditable foundation for enterprise intelligence.

Problem

• hallucinations
• multi-hop reasoning
• n-ary reasoning
• heterogeneous enterprise systems
• semantic grounding

Innovation

• Hypergraph Ontology
• Enterprise Agentic Reasoning
• Multi-hop n-ary Reasoning
• Evidence-driven Orchestration
• Auditable LLM Execution