PanoramaRAG: Enabling Consistent Global Topic Awareness in Graph-Based RAG

Graph-based Retrieval-Augmented Generation (RAG), which models relationships between fine-grained semantic units as a graph, effectively facilitates multi-hop reasoning to enhance large language model generation. However, its design focuses on local relationships, resulting in suboptimal performance for tasks that require global context, and the separation of query refinement from indexing limits the system's ability to capture high-level implicit relationships within the graph. This paper proposes a Panorama-guided RAG paradigm (PanoramaRAG) that integrates a light yet comprehensive "panorama" of the corpus to guide all stages of the retrieval process. This hub bridges the knowledge graph, language models, and queries in a computationally efficient manner, applicable to both open-source and closed-source models. Experimental results demonstrate that our method exhibits strong performance across five datasets and a variety of tasks. Our code is available at https://github.com/ Deng-Dean/PanoRAG . viding global semantic guidance without substantial computational overhead. By abstracting core content through hierarchical keywords, the panorama is seamlessly compatible with both keyword-based indexing and query expansion. We integrate this global perspective throughout the entire retrieval lifecycleencompassing offline indexing, online query construction, and document retrieval. Specifically, we introduce the following technical innovations: • PanoramaRAG: A novel RAG framework with a panoramic view of the corpus, which integrates global semantics into both the indexing and retrieval stages at minimal computational cost. By embedding corpus-level structure into retrieval-aware representations, PanoramaRAG improves semantic consistency without sacrificing efficiency. • Panorama-aware keyword construction: A method for refining queries and keywords conditioned on the panorama, ensuring that keywords consider the full context of the data before retrieval, while not limiting the types of large models (open-source or closedsource) that can be used. • Panorama-aware document retrieval: A technique that utilizes a panorama-enhanced graph and expansion mechanism to perform efficient retrieval. It improves both retrieval and generation effectiveness without adding significant overhead, leading to a notable improvement in retrieval performance.