Chapter R10-C02 · Retrieval and grounded knowledge systems
Ranking and grounded answers
Test whether added retrieval complexity improves a fixed task; Distinguish relevant passages from claims they actually support.
Units in this chapter
- R10-U03 Hybrid retrieval and reranking Test whether added retrieval complexity improves a fixed task
- R10-U04 Citations, conflicts, and abstention Distinguish relevant passages from claims they actually support
Reading that connects these units
Reading
- Sentence Transformers documentation Embeddings, semantic search, retrieval, and reranking.About this reference (W19)
- Sentence Transformers Inspect semantic search and reranking examples.About this reference (G08)
- Evaluation best practices Evaluation datasets, criteria, comparisons, and ongoing regression checks.About this reference (W17)
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Original retrieval/generation formulation; it does not guarantee current system behavior.About this reference (P02)
- AI Engineering Application patterns, evaluation, and system-level tradeoffs; select AI Engineering on the author page.About this reference (B09)
Selected, not required. Reviewed 2026-11-17.