Q65 — AWS AIF-C01 Ch.1

Question 65 of 100 | ← Chapter 1

When fine-tuning a foundation model (FM), which option is a key benefit of continued pre-training?

• A. Helps reduce model complexity
• B. Improves model performance over time ✓
• C. Reduces training time requirements
• D. Optimizes model inference time

Correct Answer: B. Improves model performance over time

Explanation

This question tests understanding of continued pre-training benefits. Continued pre-training involves further training a pre-trained FM on new, domain-relevant data—enabling it to absorb emerging patterns, terminology, and knowledge, thereby incrementally improving task performance over time. It does not inherently reduce complexity (Option A), shorten training duration (Option C), or accelerate inference (Option D). Therefore, Option B—improving model performance over time—is the primary and well-documented benefit.