Harnessing Diverse Perspectives: A Multi-Agent Framework for Enhanced Error Detection in Knowledge Graphs

2025-01-27

AI Summary

Existing knowledge graph (KG) error detection methods suffer from insufficient local structural modeling and opaque decision-making processes, hindering their reliability in industrial deployment. To address these limitations, we propose MAKGED, a multi-agent collaborative KG error detection framework. MAKGED introduces a novel LLM-based, multi-perspective subgraph-aware collaboration paradigm: it fuses fine-grained bidirectional subgraph embeddings with LLM query embeddings to instantiate four specialized agents; these agents perform structure-aware, adaptive detection via iterative, interpretable negotiation. The framework ensures decision interpretability, role specialization, and domain customizability. Extensive experiments on FB15K and WN18RR demonstrate significant improvements over state-of-the-art methods—achieving substantial gains in both accuracy and robustness. The source code and datasets are publicly released.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Multiagent Systems: Teamwork

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: Agentic search

Abstract

Knowledge graphs are widely used in industrial applications, making error detection crucial for ensuring the reliability of downstream applications. Existing error detection methods often fail to effectively leverage fine-grained subgraph information and rely solely on fixed graph structures, while also lacking transparency in their decision-making processes, which results in suboptimal detection performance. In this paper, we propose a novel Multi-Agent framework for Knowledge Graph Error Detection (MAKGED) that utilizes multiple large language models (LLMs) in a collaborative setting. By concatenating fine-grained, bidirectional subgraph embeddings with LLM-based query embeddings during training, our framework integrates these representations to produce four specialized agents. These agents utilize subgraph information from different dimensions to engage in multi-round discussions, thereby improving error detection accuracy and ensuring a transparent decision-making process. Extensive experiments on FB15K and WN18RR demonstrate that MAKGED outperforms state-of-the-art methods, enhancing the accuracy and robustness of KG evaluation. For specific industrial scenarios, our framework can facilitate the training of specialized agents using domain-specific knowledge graphs for error detection, which highlights the potential industrial application value of our framework. Our code and datasets are available at https://github.com/kse-ElEvEn/MAKGED.

Problem

• Knowledge Graph
• Error Detection
• Local Graph Utilization

Innovation

• MAKGED
• Multi-model Collaboration
• Error Detection in Knowledge Graphs