https://nxfuga.com/tags/rag/ Recent content in RAG on Popular AI Hugo -- gohugo.io en-us Mon, 20 Apr 2026 00:00:00 +0000 https://nxfuga.com/posts/note-3b7d6e05a1/ Mon, 20 Apr 2026 00:00:00 +0000 https://nxfuga.com/posts/note-3b7d6e05a1/ <h2 id="rag-quality-retrieval-first-generation-second">RAG Quality: Retrieval First, Generation Second </h2><p>When answers sound confident but conflict with your documents, the usual mistake is to tune the model before you measure retrieval. In most production RAG stacks, <strong>bad chunks</strong> or <strong>wrong ranks</strong> explain more failures than “the LLM forgot.”</p> <p>Start with the evidence trail. Log the top <em>k</em> chunk IDs, scores, and short previews for every query (with redaction for PII). If the right passage never appears in those <em>k</em> results, no amount of prompting fixes that—you need better chunking, metadata filters, embeddings refresh, or hybrid sparse+dense search.</p> <p>When retrieval looks sane and answers still drift, tighten the contract: require citations, forbid facts not present in retrieved text, and surface “not enough context” as an explicit outcome instead of guessing.</p> <p>This keeps iteration honest: you upgrade ingestion and retrieval with numbers, and you treat generation as the last mile, not the first knob.</p>