Aggregate Evaluation Metric

A method in AI model evaluation that consolidates the pass rates of all test cases into a single numerical value. While it provides an overview of overall performance, it may obscure concentrated failure patterns in specific subgroups or scenarios.