Spring AI Modular RAG
Use Jev to filter and rerank retrieved passages before a Spring AI answer.
Added to Jevfast
How it uses Jev
The creator links a Spring AI RAG example where Jev filters passages for injection, premise conflict, relevance and answer evidence, then reranks survivors before the LLM answers. This is a source-based description of the linked article; the exact API payload was not verified for this post-only entry.
Why it is interesting
The example separates broad retrieval from answer evidence. It retains conflicting evidence deliberately and reports keeping two of six retrieved chunks, despite a topK(3) cap. This is one worked example, not a general performance claim.