Evaluation & safety · Glossary term
What is Exact Match (EM)?
A metric that counts an output as correct only when its normalized representation exactly equals an accepted reference answer.
Why does Exact Match (EM) matter?
It is deterministic and easy to audit for tasks with one canonical answer, but it exposes no partial credit.
Exact Match (EM) in practice
Define normalization and all accepted references before evaluation, then pair exact match with task-specific checks when several outputs can be valid.
What is the common confusion about Exact Match (EM)?
A low exact-match score can reflect harmless formatting differences, while a matching string can still be unsupported or unsafe in context.
Learn Exact Match (EM) in the course
Lessons that name Exact Match (EM) in a title or section
- Multimodal Evaluation
Training is half the loop. The other half is measurement. This lesson builds three evaluation surfaces from primitives: image-caption retrieval reported as R@1, R@5, R@10; visual question answering…
Covered in Phase 19: Capstone Projects.
Related terms
- ROUGEA family of metrics that compares generated text with reference text using units such as n-gram overlap or longest common subsequence.
- Eval SetA versioned collection of inputs, expected properties, scoring rules, and metadata used to measure an AI system against a defined…
- Structured OutputModel output constrained or validated against a machine-readable schema so application code can consume fields without parsing free-form…
- Pass@kAcross a task set, the fraction of tasks for which at least one of k sampled candidates passes a defined correctness test.
- Benchmark ContaminationOverlap or information leakage between evaluation examples and data used to pretrain, tune, prompt, select, or otherwise improve the…
Sources
More terms in Evaluation & safety
This entry comes from glossary/terms.md on GitHub. Browse all 250 glossary terms.