Trans-RAG: Query-Centric Vector Transformation for Secure Cross-Organizational Retrieval

TL;DR

Trans-RAG introduces a novel vector transformation approach enabling secure, efficient cross-organizational retrieval without decryption, maintaining high accuracy and strong security separation.

Contribution

It proposes a new vector space language paradigm with query-centric transformations for secure, efficient cross-organizational retrieval in RAG systems.

Findings

Near-orthogonal vector spaces with 89.90° angular separation.

99.81% security isolation rate.

Only 3.5% decrease in retrieval accuracy across experiments.

Abstract

Retrieval Augmented Generation (RAG) systems deployed across organizational boundaries face fundamental tensions between security, accuracy, and efficiency. Current encryption methods expose plaintext during decryption, while federated architectures prevent resource integration and incur substantial overhead. We introduce Trans-RAG, implementing a novel vector space language paradigm where each organization's knowledge exists in a mathematically isolated semantic space. At the core lies vector2Trans, a multi-stage transformation technique that enables queries to dynamically "speak" each organization's vector space "language" through query-centric transformations, eliminating decryption overhead while maintaining native retrieval efficiency. Security evaluations demonstrate near-orthogonal vector spaces with 89.90{\deg} angular separation and 99.81% isolation rates. Experiments across 8 retrievers, 3 datasets, and 3 LLMs show minimal accuracy degradation (3.5% decrease in nDCG@10) and significant efficiency improvements over homomorphic encryption.