[2603.25374] Supercharging Federated Intelligence Retrieval

Computer Science > Information Retrieval arXiv:2603.25374 (cs) [Submitted on 26 Mar 2026] Title:Supercharging Federated Intelligence Retrieval Authors:Dimitris Stripelis, Patrick Foley, Mohammad Naseri, William Lindskog-Münzing, Chong Shen Ng, Daniel Janes Beutel, Nicholas D. Lane View a PDF of the paper titled Supercharging Federated Intelligence Retrieval, by Dimitris Stripelis and 6 other authors View PDF HTML (experimental) Abstract:RAG typically assumes centralized access to documents, which breaks down when knowledge is distributed across private data silos. We propose a secure Federated RAG system built using Flower that performs local silo retrieval, while server-side aggregation and text generation run inside an attested, confidential compute environment, enabling confidential remote LLM inference even in the presence of honest-but-curious or compromised servers. We also propose a cascading inference approach that incorporates a non-confidential third-party model (e.g., Amazon Nova) as auxiliary context without weakening confidentiality. Comments: Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL); Cryptography and Security (cs.CR); Machine Learning (cs.LG) MSC classes: 68P20, 68T05, 62M45, 68P25, 68T50, 68T10 ACM classes: H.3.3; I.2.7 Cite as: arXiv:2603.25374 [cs.IR]   (or arXiv:2603.25374v1 [cs.IR] for this version)   https://doi.org/10.48550/arXiv.2603.25374 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submiss...