Agentic-KGR: Co-evolutionary Knowledge Graph Construction through Multi-Agent Reinforcement Learning
This addresses the issue of coverage gaps and obsolescence in knowledge bases for LLMs, offering a novel method for dynamic knowledge integration, though it appears incremental as it builds on existing RL and KG techniques.
The paper tackles the problem of static knowledge bases limiting knowledge-enhanced LLMs by proposing Agentic-KGR, a framework for co-evolutionary knowledge graph construction using multi-agent RL, which achieves substantial improvements in knowledge extraction and downstream QA tasks with significant gains in accuracy and coverage.
Current knowledge-enhanced large language models (LLMs) rely on static, pre-constructed knowledge bases that suffer from coverage gaps and temporal obsolescence, limiting their effectiveness in dynamic information environments. We present Agentic-KGR, a novel framework enabling co-evolution between LLMs and knowledge graphs (KGs) through multi-round reinforcement learning (RL). Our approach introduces three key innovations: (1) a dynamic schema expansion mechanism that systematically extends graph ontologies beyond pre-defined boundaries during training; (2) a retrieval-augmented memory system enabling synergistic co-evolution between model parameters and knowledge structures through continuous optimization; (3) a learnable multi-scale prompt compression approach that preserves critical information while reducing computational complexity through adaptive sequence optimization. Experimental results demonstrate substantial improvements over supervised baselines and single-round RL approaches in knowledge extraction tasks. When integrated with GraphRAG, our method achieves superior performance in downstream QA tasks, with significant gains in both accuracy and knowledge coverage compared to existing methods.