label-errors

#计算机科学#The toolkit to test, validate, and evaluate your models and surface, curate, and prioritize the most valuable data for labeling.

#数据仓库#🛠️ Corrected Test Sets for ImageNet, MNIST, CIFAR, Caltech-256, QuickDraw, IMDB, Amazon Reviews, 20News, and AudioSet

#计算机科学#AQuA: A Benchmarking Tool for Label Quality Assessment, NeurIPS'23 D&B

#计算机科学#AQuA: A Benchmarking Tool for Label Quality Assessment, NeurIPS'23 D&B