context precision
An eval metric for RAG retrieval: whether the relevant chunks were ranked above the irrelevant ones in what was retrieved.
In more detail
Low precision means the prompt is full of noise that can distract the model. Ragas computes it from judged relevance of each retrieved chunk, weighted by rank.
Related terms
- context recall — An eval metric for RAG retrieval: the share of the facts a reference answer needs that appear in the retrieved context.
- faithfulness — An eval metric for RAG: the share of claims in an answer that the retrieved context supports.
- Ragas — An open-source library of evaluation metrics for RAG pipelines and agents, such as faithfulness, context precision, context recall and tool call accuracy.
Taught in
- Evaluating RAG And Agents — AI Engineering: LLM Apps, RAG & Agents