AI engineering glossary · Evaluation metrics
What is EM (exact match)?
Exact match (also called EM): The fraction of predictions that are identical to a reference answer, usually after normalising case, punctuation and spacing. It is strict and unambiguous, which suits questions with a single short answer, but it gives no credit for correct answers worded differently.
Exact match: an example
On 200 trivia questions, a model whose answer string matches the reference on 130 gets an exact match score of 65%.
Related terms
More Evaluation metrics terms
- Expected calibration error
- Expected reciprocal rank
- Explained variance score
- F-beta score
- F1 score
- False negative
- False negative rate
- False positive
Where you learn it
Exact match is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.