The finding that matters: generative engines retrieve pages using model-rewritten sub-queries, not the original user query, and those retrieved results dictate citations far more than traditional SEO signals do. For publishers, optimizing for traditional search engines misses what actually matters. For builders of generative engines, this highlights how internal retrieval and query rewriting shape the web's information ecosystem. Content matching question phrasing helps, but it's secondary.
RAG's weak point is that it assumes evidence sits in a database waiting to be found. RECAST flips this: evidence is computed from multiple sources through routing decisions. This matters for builders working on complex tasks (financial analysis, policy research, scientific reasoning) where the answer lives in the pattern across sources, not in any single item. Early signal on a real problem in the RAG stack.
This breaks an assumption most RAG systems are built on. If semantic identity requires computation over context, then your embeddings are not sufficient for retrieval. The implication: either you're getting lower quality matches than you think, or you need to rethink how you index and query. The finding is clean and holds across multiple model families, so it's not a one-off artifact.
RAG in production is mostly dark: you ship it and hope. Re:CAP solves the real problem: you can't label a billion-document corpus but you can ask whether your retriever is missing obvious stuff. The 9-29% recovery gap against BM25 is substantial. If you have a RAG pipeline in production, add this audit loop to your monitoring before someone else's system gives you the answer yours can't find.
This is a serious finding: RLVR training for correct answers does not buy citation faithfulness, and the unfaithful citation rate is quantified at scale. If your RAG system is user-facing and cites sources, this says you cannot rely on reward-based training to fix false attribution. Citation accuracy needs explicit measurement and training.