Cross-Entropy Loss measures the difference between two probability distributions, often used in classification tasks. It penalizes heavily when the predicted probability for the correct class is low. This loss function is a standard choice for multi-class problems with softmax outputs.