BenchmarkEMNIST数据集
EMNIST
An extension of the MNIST dataset that contains handwritten letters as well as digits. It is commonly used in research on image classification and character recognition.
Timeline (last 90 days)
Oct 19
The experiment uses the EMNIST letters and MNIST digits datasets, constructing a distribution shift that is disjoint from the support of the original distribution yet completely blinds the CTM by conditionally resampling the EMNIST letters so that their conformal scores exactly match those of the MNIST digits.
Unverified50%