The benchmarks section lists all benchmarks using a given dataset or any of
its variants. We use variants to distinguish between results evaluated on
slightly different versions of the same dataset. For example, ImageNet 32⨉32
and ImageNet 64⨉64 are variants of the ImageNet dataset.
The largest and cleanest face recognition dataset Glint360K,
which contains 17,091,657 images of 360,232 individuals, baseline models trained on Glint360K can easily achieve state-of-the-art performance.