SmartFAQs.ai
Back to Learn
Intermediate

Recall

The ratio of relevant documents successfully retrieved to the total number of relevant documents available in the knowledge base. In RAG pipelines, it quantifies the system's ability to minimize 'false negatives' by ensuring all contextually necessary information is provided to the LLM.

Definition

The ratio of relevant documents successfully retrieved to the total number of relevant documents available in the knowledge base. In RAG pipelines, it quantifies the system's ability to minimize 'false negatives' by ensuring all contextually necessary information is provided to the LLM.

Disambiguation

Measures the 'completeness' of retrieval (minimizing omissions) rather than the 'purity' of the results (precision).

Visual Metaphor

"A giant safety net at a circus designed to catch a performer regardless of where they fall, prioritizing total coverage over a narrow target."

Key Tools
RAGASDeepEvalTruLensArize PhoenixLangSmith
Related Connections

Conceptual Overview

The ratio of relevant documents successfully retrieved to the total number of relevant documents available in the knowledge base. In RAG pipelines, it quantifies the system's ability to minimize 'false negatives' by ensuring all contextually necessary information is provided to the LLM.

Disambiguation

Measures the 'completeness' of retrieval (minimizing omissions) rather than the 'purity' of the results (precision).

Visual Analog

A giant safety net at a circus designed to catch a performer regardless of where they fall, prioritizing total coverage over a narrow target.

Related Articles