Retrieval & generation · Glossary term
What is Recall@K?
For one query, Recall@K is `|relevant items intersecting the top k| / |relevant items|`. A dataset score aggregates those per-query values under a stated rule.
Why does Recall@K matter?
It tells you whether a retrieval stage supplies downstream generation or reranking with enough relevant candidates.
Recall@K in practice
Define relevance judgments, k, the aggregation method, and a policy for queries with no judged relevant items, then inspect queries with zero recalled evidence.
What is the common confusion about Recall@K?
High Recall@K does not mean the top result is good, the ranking is well ordered, or the final answer is grounded. Queries with no relevant items require an explicit exclusion or assigned-value policy because the denominator is zero.
Learn Recall@K in the course
Lessons that name Recall@K in a title or section
- Image Retrieval & Metric Learning
A retrieval system ranks candidates by a distance in embedding space. Metric learning is the discipline of shaping that space so the distances mean what you want.
- Multimodal Evaluation
Training is half the loop. The other half is measurement. This lesson builds three evaluation surfaces from primitives: image-caption retrieval reported as R@1, R@5, R@10; visual question answering…
- Chunking Strategies, Compared
Chunking decides what your retriever can ever surface. Get the boundaries wrong and no embedding model, no reranker, no LLM can repair the damage downstream.
- RAG Evaluation: Precision, Recall, MRR, nDCG, Faithfulness, Answer Relevance
If you cannot grade your retrieval and your answer at the same time, you cannot ship the system. The two are not the same metric and the same prompt fails on different axes.
Covered in Phase 04: Computer Vision and Phase 19: Capstone Projects.
Related terms
- Precision & RecallPrecision asks how many flagged items were correct; recall asks how many relevant items were found.
- Eval SetA versioned collection of inputs, expected properties, scoring rules, and metadata used to measure an AI system against a defined…
- RerankerA second-stage model or scoring function that reorders a small candidate set using a richer comparison between the query and each candidate.
- Approximate Nearest Neighbor (ANN)A search method that returns vectors likely to be among the nearest to a query without exhaustively comparing the query with every stored…
- HNSWAn approximate-nearest-neighbor index that organizes vectors in layered proximity graphs and searches from coarse upper layers toward…
Sources
More terms in Retrieval & generation
This entry comes from glossary/terms.md on GitHub. Browse all 250 glossary terms.