Privacy Without Losing Place: A Paradigm for Private Retrieval in Spatial RAGs

TL;DR

This paper proposes PAS, a structured spatial privacy mechanism for RAG systems that balances location privacy with retrieval performance, using relative anchor encoding instead of direct perturbation.

Contribution

PAS introduces a novel relative anchor encoding scheme for spatial privacy, enabling effective privacy-utility trade-offs in RAG systems without relying on traditional noise-based methods.

Findings

PAS achieves 370-400m adversarial location error with strong privacy guarantees.

Retrieval performance remains over 50% of baseline despite privacy-preserving modifications.

Downstream generation quality remains robust under PAS, showing LLMs can compensate for spatial retrieval imperfections.

Abstract

This work introduces PAS -- Privacy Anchor Substitution, a structured mechanism for enabling user location privacy in spatial retrieval-augmented generation (RAG) systems. Unlike conventional differential privacy methods that directly perturb user locations, PAS represents location with relative anchor encoding consisting of an anchor, direction bin, and distance bin, allowing seamless integration with modern RAG pipelines. We evaluate PAS on a synthetic urban dataset and show that it achieves impressive coarse privacy guarantees, with approximately 370-400m adversarial location error, while retaining more than half of the baseline retrieval performance. Despite the slight drop in retrieval performance, the downstream generation quality under PAS remains comparatively robust, indicating that large language models can compensate for imperfect spatial retrieval. Furthermore, we provide empirical analysis showing that PAS exhibits non-monotonic privacy-utility relationship with respect to privacy parameters. We attribute this to geometric bias induced by anchor discretization, making it different from continuous noise mechanisms such as geo-indistinguishability. Our results show that structured spatial representations offer a practical approach to privacy in location based reasoning in RAG systems.