evaluation benchmarks

Evaluation benchmarks are reference datasets and metrics used to objectively compare the performance of different models or approaches within a specific field, facilitating advancements by providing a standardized way to assess progress.