AI-103 Question Preview

Safe Public Question Preview 3

Only the demo-safe public version is shown here. Correct answers, full explanations, internal IDs, and paid-only content stay protected.

Multi SelectImplement generative AI and agentic solutions
Question 3. Your organization is optimizing an existing generative AI application in Azure AI Foundry where users report that responses sometimes contain hallucinations and lack sufficient reasoning depth. You need to improve response quality through model tuning and evaluation strategies. Which combination of techniques should you implement? (Select two.)
A. Use reasoning-focused evaluations, structured prompting, and reflection or self-critique patterns to improve answer quality.
B. Configure groundedness, relevance, and fabrication-related evaluations to measure response quality over time.
C. Increase temperature to its maximum value for all responses.
D. Disable safety filters to generate more detailed responses.
E. Replace the model with a smaller language model without conducting evaluations.
Explanation is locked. Start the demo or unlock full access to review explanations inside the protected test engine.
Open safe question previewReport Question
Start Interactive DemoBuy Full Access

Related Safe Previews

Question 1Question 2Question 4Question 5All Preview Questions

FAQ

Does this page show the answer?

No. Answers and explanations are protected inside the private test engine.