Search papers, labs, and topics across Lattice.
NaviRAG is introduced to address the limitations of flat retrieval in RAG systems by structuring knowledge documents hierarchically and using an LLM agent to actively navigate this structure. The agent iteratively identifies information gaps and retrieves content from appropriate granularity levels, enabling conditional retrieval and dynamic synthesis. Experiments on long-document QA benchmarks demonstrate that NaviRAG improves both retrieval recall and end-to-end answer performance compared to conventional RAG baselines.
Forget flat retrieval: NaviRAG's hierarchical knowledge navigation boosts RAG performance by dynamically retrieving information across different levels of granularity.
Retrieval-augmented generation (RAG) typically relies on a flat retrieval paradigm that maps queries directly to static, isolated text segments. This approach struggles with more complex tasks that require the conditional retrieval and dynamic synthesis of information across different levels of granularity (e.g., from broad concepts to specific evidence). To bridge this gap, we introduce NaviRAG, a novel framework that shifts from passive segment retrieval to active knowledge navigation. NaviRAG first structures the knowledge documents into a hierarchical form, preserving semantic relationships from coarse-grained topics to fine-grained details. Leveraging this reorganized knowledge records, a large language model (LLM) agent actively navigates the records, iteratively identifying information gaps and retrieving relevant content from the most appropriate granularity level. Extensive experiments on long-document QA benchmarks show that NaviRAG consistently improves both retrieval recall and end-to-end answer performance over conventional RAG baselines. Ablation studies confirm performance gains stem from our method's capacity for multi-granular evidence localization and dynamic retrieval planning. We further discuss efficiency, applicable scenario, and future directions of our method, hoping to make RAG systems more intelligent and autonomous.
