Inference Scaled GraphRAG: Improving Multi Hop Question Answering on Knowledge Graphs

TL;DR

Inference-Scaled GraphRAG enhances large language models' multi-hop reasoning on knowledge graphs by applying inference-time compute scaling, leading to significant improvements in question answering performance.

Contribution

It introduces a novel inference-time scaling framework combining sequential and parallel methods to improve graph reasoning in LLMs.

Findings

Significant performance gains on the GRBench benchmark.

Outperforms traditional GraphRAG and graph traversal baselines.

Demonstrates effectiveness of inference-time scaling for structured knowledge reasoning.

Abstract

Large Language Models (LLMs) have achieved impressive capabilities in language understanding and generation, yet they continue to underperform on knowledge-intensive reasoning tasks due to limited access to structured context and multi-hop information. Retrieval-Augmented Generation (RAG) partially mitigates this by grounding generation in retrieved context, but conventional RAG and GraphRAG methods often fail to capture relational structure across nodes in knowledge graphs. We introduce Inference-Scaled GraphRAG, a novel framework that enhances LLM-based graph reasoning by applying inference-time compute scaling. Our method combines sequential scaling with deep chain-of-thought graph traversal, and parallel scaling with majority voting over sampled trajectories within an interleaved reasoning-execution loop. Experiments on the GRBench benchmark demonstrate that our approach significantly improves multi-hop question answering performance, achieving substantial gains over both traditional GraphRAG and prior graph traversal baselines. These findings suggest that inference-time scaling is a practical and architecture-agnostic solution for structured knowledge reasoning with LLMs