A team prepares data for instruction fine-tuning. Some examples contain conflicting target answers, and others include customer information that the team is not authorized to use for training. Which actions should it take before training? Select 2 answers.

1. Confirm data-use rights and privacy requirements, and remove or appropriately transform records that cannot be used.
2. Curate consistent, representative instruction-response examples and assess label or target-answer quality.
3. Use the final evaluation set as training data to ensure the reported evaluation score rises.
4. Retain contradictory examples unchanged and rely on a higher inference temperature to fix training quality.
5. Include unauthorized records because more data always compensates for governance issues.