Topology-Aware Reasoning over Incomplete Knowledge Graph with Graph-Based Soft Prompting

2026-04-14

AI Summary

This work addresses the fragility of multi-hop knowledge graph question answering (KBQA) under incomplete graphs, where performance degrades due to reliance on explicit path traversal. To overcome this limitation, the authors propose a subgraph-level soft prompting framework that shifts reasoning from node paths to structured subgraphs. Specifically, query-relevant subgraphs are first extracted, and their topological structures are encoded via graph neural networks to generate soft prompts that guide large language models (LLMs) in capturing rich contextual information for answer entity identification. The approach introduces, for the first time, a subgraph-level soft prompting mechanism and integrates a two-stage collaborative architecture combining lightweight and powerful LLMs, significantly reducing computational overhead while enhancing robustness to missing relations. The method achieves state-of-the-art results on three out of four multi-hop KBQA benchmarks, substantially outperforming existing approaches.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Question Answering
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Large Language Models (LLMs) have shown remarkable capabilities across various tasks but remain prone to hallucinations in knowledge-intensive scenarios. Knowledge Base Question Answering (KBQA) mitigates this by grounding generation in Knowledge Graphs (KGs). However, most multi-hop KBQA methods rely on explicit edge traversal, making them fragile to KG incompleteness. In this paper, we proposed a novel graph-based soft prompting framework that shifts the reasoning paradigm from node-level path traversal to subgraph-level reasoning. Specifically, we employ a Graph Neural Network (GNN) to encode extracted structural subgraphs into soft prompts, enabling LLM to reason over richer structural context and identify relevant entities beyond immediate graph neighbors, thereby reducing sensitivity to missing edges. Furthermore, we introduce a two-stage paradigm that reduces computational cost while preserving good performance: a lightweight LLM first leverages the soft prompts to identify question-relevant entities and relations, followed by a more powerful LLM for evidence-aware answer generation. Experiments on four multi-hop KBQA benchmarks show that our approach achieves state-of-the-art performance on three of them, demonstrating its effectiveness. Code is available at the repository: https://github.com/Wangshuaiia/GraSP.

Problem

• Knowledge Graph Incompleteness
• Multi-hop KBQA
• Reasoning over Incomplete KG
• LLM Hallucination
• Graph-based Reasoning

Innovation

• graph-based soft prompting
• subgraph-level reasoning
• knowledge graph incompleteness
• two-stage KBQA
• GNN-LLM integration