Researchers have developed VDGR-RAG, a unified framework for enterprise knowledge question answering. It integrates vector retrieval, directory-driven reasoning, graph traversal, and iterative reflection to improve accuracy over hierarchical document structures. The system constructs a Hierarchical Heterogeneous Knowledge Graph from document chunks and employs four tools: routing, multi-route retrieval, backtracking, and dynamic reflection. Experiments on wireless domains show that VDGR-RAG outperforms RAG baselines in knowledge retrieval recall and QA accuracy. --- Why it matters: This matters to AI researchers because it addresses limitations in existing Retrieval-Augmented Generation approaches, which often overlook the integration of diverse retrieval strengths. The proposed framework has significant implications for improving enterprise knowledge question answering capabilities. Source: https://arxiv.org/abs/2608.07994