A legal research assistant retrieves case law from a vector store. Users report that retrieved cases are topically relevant but jurisdictionally mismatched — a California employment case is retrieved for a New York labor question. The vector store supports metadata filtering. What change addresses this with the least retrieval quality regression?

A is correct because jurisdiction is a hard categorical constraint, not a soft relevance signal. A pre-filter on metadata ensures the ANN search only considers documents matching the required jurisdiction, eliminating the class of mismatches entirely without degrading within-jurisdiction ranking quality. This is the standard pattern when a metadata attribute is a hard exclusion criterion. B is wrong because embedding jurisdiction as a text prefix causes the embedding to partially encode jurisdictional identity, which may help but does not guarantee correctness — the embedding space is continuous and a California case with many overlapping terms could still outscore a New York case. C is wrong because retrieving 50 broad candidates and post-filtering is less efficient and risks dropping below an adequate candidate count after filtering, especially for rare jurisdictions. D is wrong because using Claude as a post-retrieval filter introduces unnecessary LLM calls, adds latency and cost, and relies on model behavior rather than deterministic metadata logic for what should be a deterministic constraint.