Beyond Isolation: Multi-Agent Synergy for Improving Knowledge Graph Construction
This work addresses knowledge graph construction for AI and data science applications, presenting an incremental improvement over existing methods.
The paper tackles the problem of knowledge graph construction by challenging the solitary approach of large language models, introducing CooperKGC, a collaborative multi-agent framework that improves knowledge selection, correction, and aggregation through concurrent task handling.
This paper introduces CooperKGC, a novel framework challenging the conventional solitary approach of large language models (LLMs) in knowledge graph construction (KGC). CooperKGC establishes a collaborative processing network, assembling a team capable of concurrently addressing entity, relation, and event extraction tasks. Experimentation demonstrates that fostering collaboration within CooperKGC enhances knowledge selection, correction, and aggregation capabilities across multiple rounds of interactions.