A legal research assistant uses RAG to retrieve relevant case law. Attorneys report that keyword-specific queries (searching for a specific statute number, '42 U.S.C. § 1983') return irrelevant cases that are semantically similar but do not cite the exact statute. The system uses only dense vector retrieval. What change best fixes this?

C is correct because hybrid BM25+vector retrieval addresses exactly this failure mode: BM25 excels at exact term matching (statute numbers, case citations, named parties) while vector retrieval excels at semantic similarity. RRF fusion produces a ranked list that rewards documents appearing highly in both lists, ensuring exact-match documents surface even when they are not the closest semantic neighbors. A is wrong because retrieving top-20 instead of top-5 increases recall but does not solve the precision problem — the statute-specific documents may still rank below semantically similar but irrelevant cases. B is wrong because fine-tuning the embedding model would require legal training data and compute investment, and may not adequately capture exact-match behavior for specific numeric identifiers like statute numbers — BM25 handles this natively. D is wrong because lowering the similarity threshold accepts more documents of lower relevance without specifically prioritizing exact-match documents.