FlagThis

Abstract

Dense embeddings underpin semantic search and Retrieval-Augmented Generation, yet a leaked vector store hands much of the underlying text back to whoever holds it. The modern attacks that make this possible—few-shot alignment, zero-shot inversion, unsupervised cross-space translation—all turn on the same weakness: the protected store is a _single global geometry_, and any single geometry can be aligned to a known one. We introduce Shard, a retrieval-preserving embedding transform built to remove that weak axis rather than to mask a distortion surrogate.