aiqbench analysis
This arXiv preprint proposes Federated GraphRAG, a way for LLMs to retrieve and reason over knowledge graphs held by multiple organizations without centralizing them or exposing their raw graph structures. The core ideas are a routing mechanism that uses lightweight knowledge indices to find relevant clients and a controlled-disclosure method that shares abstracted evidence rather than local graph memory. Readers should treat the reported gains over baselines as preliminary until they can check the benchmarks and privacy claims in the full paper.
Federated Graph Retrieval-Augmented Generation
arXiv:2610.30728v1 Announce Type: new Abstract: Graph Retrieval-Augmented Generation (GraphRAG) leverages centralized knowledge graphs to enhance large language models with structured knowledge retrieval. However, real-world knowledge is inherently distributed across multiple organizations rather than being centrally accessible. Privacy constraints further prohibit sharing entire local knowledge graphs, preventing GraphRAG from effectively retrieving and utilizing distributed knowledge. To this end, we propose Federated GraphRAG (FedGRAG), a unified framework for distributed graph retrieval and knowledge-enhanced generation over multiple clients with knowledge graphs. Specifically, we address two fundamental challenges in federated GraphRAG. First, the absence of a global knowledge graph makes unified retrieval impossible. To address this, we develop a Federated Knowledge Routing mechanism that efficiently identifies relevant clients through lightweight knowledge indices without exposing local graph structures. Second, local graph memory cannot be shared directly because it contains sensitive entities and knowledge information. We therefore propose a controlled-disclosure evidence construction method. It extracts retrieved evidence subgraphs and converts them into an evidence abstract, enabling evidence conversion and reconstruction on the server. Extensive experiments demonstrate that Federated GraphRAG consistently outperforms existing baselines in both retrieval and downstream generation tasks, establishing a new paradigm for deploying GraphRAG in federated environments.
Abstract:Graph Retrieval-Augmented Generation (GraphRAG) leverages centralized knowledge graphs to enhance large language models with structured knowledge retrieval. However, real-world knowledge is inherently distributed across multiple organizations rather than being centrally accessible. Privacy constraints further prohibit sharing entire local knowledge graphs, preventing GraphRAG from effectively retrieving and utilizing distributed knowledge. To this end, we propose Federated GraphRAG (FedGRAG), a unified framework for distributed graph retrieval and knowledge-enhanced generation over multiple clients with knowledge graphs. Specifically, we address two fundamental challenges in federated GraphRAG. First, the absence of a global knowledge graph makes unified retrieval impossible. To address this, we develop a Federated Knowledge Routing mechanism that efficiently identifies relevant clients through lightweight knowledge indices without exposing local graph structures. Second, local graph memory cannot be shared directly because it contains sensitive entities and knowledge information. We therefore propose a controlled-disclosure evidence construction method. It extracts retrieved evidence subgraphs and converts them into an evidence abstract, enabling evidence conversion and reconstruction on the server. Extensive experiments demonstrate that Federated GraphRAG consistently outperforms existing baselines in both retrieval and downstream generation tasks, establishing a new paradigm for deploying GraphRAG in federated environments.
Machine Learning (cs.LG)
arXiv:2610.30728 [cs.LG]
(or arXiv:2610.30728v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.30728 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 283 words
Related coverage
Effects of Transcript Compression on LLM-based Medical Misinformation Detection in Japanese YouTube Videos
arXiv · Sep 28, 2026
TRACE: Accountable Agentic Retrieval for Source Discovery in Digital Archives
arXiv · Sep 18, 2026
Trajectory-Aware Retrieval Agents for Temporal Decision- Making
arXiv · Jul 28, 2026