Fine-tuned models are often overconfident: they say 99% but are right 80% of the time. Temperature scaling (Guo et al., 2017) fixes this by dividing the logits by a single learned T > 1 before the softmax. It's fitted on a validation set and doesn't change which class is predicted.
Unlock this and every premium problem across all tracks — plus hidden-test grading, advanced AI engineering modules and new content every week.
Start 7-day free trial