A Large Language Model Based Method for Complex Logical Reasoning over Knowledge Graphs

TL;DR

This paper introduces ROG, a novel framework combining knowledge graph retrieval with large language model reasoning to improve complex logical query answering over KGs, outperforming embedding-based methods.

Contribution

The paper presents ROG, a new ensemble approach that integrates query-aware KG retrieval with LLM-based chain-of-thought reasoning for complex FOL queries.

Findings

ROG outperforms embedding-based baselines on standard benchmarks.

Significant improvements on high-complexity query types.

Demonstrates robustness of LLM-driven reasoning in KG tasks.

Abstract

Reasoning over knowledge graphs (KGs) with first-order logic (FOL) queries is challenging due to the inherent incompleteness of real-world KGs and the compositional complexity of logical query structures. Most existing methods rely on embedding entities and relations into continuous geometric spaces and answer queries via differentiable set operations. While effective for simple query patterns, these approaches often struggle to generalize to complex queries involving multiple operators, deeper reasoning chains, or heterogeneous KG schemas. We propose ROG (Reasoning Over knowledge Graphs with large language models), an ensemble-style framework that combines query-aware KG neighborhood retrieval with large language model (LLM)-based chain-of-thought reasoning. ROG decomposes complex FOL queries into sequences of simpler sub-queries, retrieves compact, query-relevant subgraphs as contextual evidence, and performs step-by-step logical inference using an LLM, avoiding the need for task-specific embedding optimization. Experiments on standard KG reasoning benchmarks demonstrate that ROG consistently outperforms strong embedding-based baselines in terms of mean reciprocal rank (MRR), with particularly notable gains on high-complexity query types. These results suggest that integrating structured KG retrieval with LLM-driven logical reasoning offers a robust and effective alternative for complex KG reasoning tasks.