Results 161 to 170 of about 14,353 (222)

AI and Measurement Concerns: Dealing with Imbalanced Data in Autoscoring

open access: yesJournal of Educational Measurement, Volume 63, Issue 1, Spring 2026.
Abstract Unbiasedness for proficiency estimates is important for autoscoring engines since the outcome might be used for future learning or placement. Imbalanced training data may lead to certain biases and lower the prediction accuracy for classification algorithms.
Yunting Liu   +3 more
wiley   +1 more source

Large language models reflect the ideology of their creators. [PDF]

open access: yesNPJ Artif Intell
Buyl M   +10 more
europepmc   +1 more source

Reply to Yu et al.: Datasets, human judges, and future directions for evaluating AI-AI bias. [PDF]

open access: yesProc Natl Acad Sci U S A
Kulveit J   +5 more
europepmc   +1 more source

Home - About - Disclaimer - Privacy