Glossary · Validity and evidence
Transport
4ValidationWhether an evaluator's measured error rates carry over across generators, slices, time, and environments. A component of the validation profile; it is what licenses the validity domain.
Defined in
Handbook · Validity under pressureThe validation profile and validity statementReplace hierarchy labels with measurements, Π = (N, E, T, G, C, D), and ship every evaluator with a falsifiable statement of scope.Related terms
- Validity domain
- The declared conditions and population over which an interpretation is justified. Outside it, a benchmark has made no justified claim.
- Validation profile
- The measured profile Π = (N, E, T, G, C, D): natural error, exploitability, transport, generator coupling, channel coupling, drift. It replaces hierarchy labels with measurements.
Where it comes up
- Notation and formula cardHandbook · Reference