Exam AWS Certified AI Practitioner AIF-C01 topic 1 question 185 discussion

✅ Fine-tuning The company wants to improve the model’s domain knowledge by providing specific documents. is fine tuning not pre-training. Pretraining refers to training a model from scratch on a large dataset to learn general features and representations. It is not focused on adapting to specific domain knowledge. Since the company already has a model and wants to improve its domain knowledge with specific documents, fine-tuning is the more suitable choice. Pretraining refers to training a model on a large general dataset before any specific task, which may not be as effective for domain-specific enhancements

The company wants to improve the model's performance on specific tasks and examples. ✅ Fine-tuning (Fine-tuning adapts a pretrained model to perform better on a specific task by training it on task-specific labeled data.) The company wants to improve the model’s domain knowledge by providing specific documents. ✅ Continued pre-training (Continued pre-training exposes the model to more unlabeled domain-specific data, enhancing its general understanding within that domain.) The company wants to retrain the model by using more unlabeled data over time. ✅ Continued pre-training (This technique helps the model learn from additional unlabeled data, often used to adapt to evolving language or domain shifts.)