CIFAR-10H
PulseAugur coverage of CIFAR-10H — every cluster mentioning CIFAR-10H across labs, papers, and developer communities, ranked by signal.
-
AI model uncertainty fails to track human disagreement on image tasks
A new study published on Hugging Face's Daily Papers investigated whether AI model uncertainty correlates with human disagreement on image labeling tasks. Researchers evaluated eight pre-trained models across three arch…
-
New conformal inference method tackles label noise in machine learning
Researchers have developed a new method for conformal inference that addresses the challenge of label noise in machine learning datasets. This adaptive technique ensures accurate uncertainty quantification and provides …
-
Soft-label training enhances classifier accuracy by using full annotation distributions
Researchers have proposed a new training method called soft-label training for supervised classifiers, which utilizes the full distribution of annotator labels rather than a single majority vote. This approach is partic…
-
New research finds temperature scaling fails on soft labels
A new research paper challenges the effectiveness of temperature scaling for model calibration, particularly when dealing with soft or distributional human labels. The study found that temperature scaling, which assumes…