In multi-tenant RAG, same-index account collusion degrades per-account DP at Θ(√k · ε_acc) for Gaussian retrieval; a new audit protocol using Merkle ledgers, ZK proofs, and six RAG-specific primitives enables quantitative (PASS, ε_audit) verification of the noise-then-select step without index or模型
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Jacobian-guided reshaping converts isotropic LDP noise into an anisotropic distribution focused on task-relevant subspaces, yielding roughly 20% utility gains on CIFAR-10-C for PrivUnit2 and PrivUnitG at ε=7.5 while keeping per-dimension privacy budgets uniform.
citing papers explorer
-
Auditing Privacy in Multi-Tenant RAG under Account Collusion
In multi-tenant RAG, same-index account collusion degrades per-account DP at Θ(√k · ε_acc) for Gaussian retrieval; a new audit protocol using Merkle ledgers, ZK proofs, and six RAG-specific primitives enables quantitative (PASS, ε_audit) verification of the noise-then-select step without index or模型
-
Jacobian-Guided Anisotropic Noise Reshaping for Enhancing Representation Utility under Local Differential Privacy
Jacobian-guided reshaping converts isotropic LDP noise into an anisotropic distribution focused on task-relevant subspaces, yielding roughly 20% utility gains on CIFAR-10-C for PrivUnit2 and PrivUnitG at ε=7.5 while keeping per-dimension privacy budgets uniform.