Vector RAG vs LLM-Compiled Wiki: A Preregistered Comparison on a Small Multi-Domain Research

In three linesPreregistered study comparing Vector RAG and LLM-compiled markdown wiki on 13 questions over 24 papers. Wiki excels at cross-paper synthesis and claim-level citation accuracy, but uses more query tokens. A decomposition-based RAG variant recovers most wiki advantages at lower LLM-token cost.

Summary generated by Claude — human-verified