ModelRefs / Citation Accuracy — AI Glossary
Citation Accuracy — AI Glossary
The share of cited passages that actually support the statement they are attached to — distinct from whether the answer is correct.
Overview
Citation accuracy measures attribution, not correctness. An answer can be factually right while citing the wrong passage, and it can cite a real retrieved chunk that does not contain the claim. Measure it separately from grounding and from answer quality: sample answers, check each claim against the passage it points to, and score support / partial support / unsupported. It is a leading trust signal in RAG, because users verify by following citations — a plausible answer with a mis-attached citation is harder to catch than an obvious error. Common mistakes: treating retrieval recall as proof of citation quality, scoring only the first citation, or accepting document-level citations when the claim needs passage-level support.
Reference details
| Topic | rag |
|---|---|
| Also known as | attribution accuracy, citation precision |
| Last reviewed | 2026-06-24 |
Related terms
Where this appears on ModelRefs
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Citation Accuracy — AI Glossary.
Frequently asked questions
What is Citation Accuracy?
The share of cited passages that actually support the statement they are attached to — distinct from whether the answer is correct.
Is Citation Accuracy the same as attribution accuracy?
Yes — attribution accuracy, citation precision are common aliases for Citation Accuracy.
What concepts are related to Citation Accuracy?
Closely related concepts include grounding, hallucination, context precision, context recall.