Glossary of AI Terminology

What Is Precision@K?

Precision@K

Precision@K measures how many of the top K retrieved results are relevant. If a retriever returns five documents and three are relevant, Precision@5 is 0.6.

Precision@K is useful when context window space is limited or irrelevant documents confuse the model. High recall with low precision can still hurt RAG because the right answer is buried among distracting context.

Bi-weekly AI Research Paper Readings

Stay on top of emerging trends and frameworks.