ModelRefs / Corrective RAG (CRAG) — AI Glossary

Corrective RAG (CRAG) — AI Glossary

A RAG method evaluating retrieved document relevance and triggering web search correction when documents are irrelevant or ambiguous.

Overview

CRAG (Yan et al. 2024) adds a lightweight retrieval evaluator that scores retrieved documents as correct, ambiguous, or incorrect. On incorrect/ambiguous, CRAG queries the web for fresher information and applies a decompose-then-recompose knowledge refinement step. Improves factual accuracy on KILT and PopQA without full agentic overhead.

Reference details

Topic	rag
Last reviewed	2026-06-24

Commonly confused with

Grades retrieved documents for relevance and, when they fail, goes somewhere else — typically web search — instead of generating from poor context. The trigger is what separates it: Self-RAG reflects on its own generation, agentic RAG decides retrieval strategy up front, and this reacts to a measured retrieval failure. It presumes a fallback source exists, which is a real deployment constraint.

Primary source

Yan et al., Corrective Retrieval Augmented Generation (arXiv)

Continue your research

Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Corrective RAG (CRAG) — AI Glossary.

Frequently asked questions

What is Corrective RAG (CRAG)?