@inproceedings{NEURIPS DATASETS AND BENCHMARKS2021_f2217062, author = {Northcutt, Curtis and Athalye, Anish and Mueller, Jonas}, booktitle = {Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks}, editor = {J. Vanschoren and S. Yeung}, pages = {}, publisher = {Curran}, title = {Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks}, url = {https://datasets-benchmarks-proceedings.neurips.cc/paper_files/paper/2021/file/f2217062e9a397a1dca429e7d70bc6ca-Paper-round1.pdf}, volume = {1}, year = {2021} }