Leveraging LLM-GNN Integration for Open-World Question Answering over Knowledge Graphs

TL;DR

GLOW is a hybrid system combining GNNs and LLMs for open-world question answering over incomplete knowledge graphs, enabling joint reasoning without retrieval or fine-tuning.

Contribution

It introduces GLOW, a novel GNN-LLM hybrid approach for open-world KGQA, and GLOW-BENCH, a benchmark for evaluating generalization on incomplete KGs.

Findings

GLOW outperforms existing systems on standard benchmarks.

Achieves up to 53.3% accuracy and 38% average improvement.

Demonstrates effective joint reasoning over symbolic and semantic signals.

Abstract

Open-world Question Answering (OW-QA) over knowledge graphs (KGs) aims to answer questions over incomplete or evolving KGs. Traditional KGQA assumes a closed world where answers must exist in the KG, limiting real-world applicability. In contrast, open-world QA requires inferring missing knowledge based on graph structure and context. Large language models (LLMs) excel at language understanding but lack structured reasoning. Graph neural networks (GNNs) model graph topology but struggle with semantic interpretation. Existing systems integrate LLMs with GNNs or graph retrievers. Some support open-world QA but rely on structural embeddings without semantic grounding. Most assume observed paths or complete graphs, making them unreliable under missing links or multi-hop reasoning. We present GLOW, a hybrid system that combines a pre-trained GNN and an LLM for open-world KGQA. The GNN predicts top-k candidate answers from the graph structure. These, along with relevant KG facts, are serialized into a structured prompt (e.g., triples and candidates) to guide the LLM's reasoning. This enables joint reasoning over symbolic and semantic signals, without relying on retrieval or fine-tuning. To evaluate generalization, we introduce GLOW-BENCH, a 1,000-question benchmark over incomplete KGs across diverse domains. GLOW outperforms existing LLM-GNN systems on standard benchmarks and GLOW-BENCH, achieving up to 53.3% and an average 38% improvement. GitHub code and data are available.