跨组织检索中实现安全与效率兼顾的向量转换方法
Trans-RAG: Query-Centric Vector Transformation for Secure Cross-Organizational Retrieval
- Query-centric向量变换让查询动态适配各方语义空间
- 向量空间夹角达89.90°，隔离率99.81%，安全性高
- 相比同态加密提升效率，准确率损失仅3.5%（nDCG@10）
跨组织部署的检索增强生成系统面临安全、精度与效率之间的根本矛盾。现有加密方法在解密时暴露明文，而联邦架构阻碍资源融合并带来显著开销。本文提出Trans-RAG，采用新型向量空间语言范式，使各组织知识存在于数学上隔离的语义空间中。核心为query-centric的多阶段变换技术——vector2Trans，使查询能动态‘翻译’为各组织的向量空间‘语言’，消除解密开销的同时保持原生检索效率。安全评估显示向量空间近正交，平均夹角89.90°，隔离率达99.81%。在8个检索器、3个数据集、3个大模型上的实验表明，准确率仅下降3.5%（nDCG@10），显著优于同态加密。
原文摘要 · Abstract (English)
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° 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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。