A demand-prediction model was accurate at launch. Months later, customer behavior changes and prediction errors increase. Which practices support ongoing production quality rather than assuming the original evaluation remains valid? Select 2 answers.

1. Evaluate a retrained candidate on representative held-out data before replacing the deployed model.
2. Automatically deploy every retraining result without evaluation because newer models are always better.
3. Use the original training accuracy as the only metric for the lifetime of the application.
4. Delete recent outcome labels so the original model cannot be contradicted by new evidence.
5. Monitor input distributions and prediction quality using suitable current data and available outcome labels.