שאלה | Why is nested cross-validation considered more reliable for model evaluation?

Why is nested cross-validation considered more reliable for model evaluation?

		Because it provides a biased estimate of the model's performance.
		Because it requires less computational resources compared to traditional cross-validation.
		Because it focuses solely on hyperparameter tuning without evaluating the model's generalization ability.
		Because it accounts for both hyperparameter tuning and model evaluation on unseen data, resulting in a more accurate performance estimate.