Preparing Practice Exam…
Please wait a moment while the system generates the exam questions, answers, and study session.
Please wait a moment while the system generates the exam questions, answers, and study session.
Question 26 of 44
Free Preview · 44 of 146 · questions 1–44 of 146 are free. Sign up free to sync your progress across devices.
Select one answer.
Hotspot Solution:
| Prompting Claim | Sound Practice? |
|---|---|
| Zero-shot prompting is a reasonable baseline for a clearly defined closed-set classifier | Yes |
| Few-shot examples containing exact field names improve extraction consistency | Yes |
| Chain-of-thought should be enabled unconditionally for every routine task regardless of latency | No |
| Even-handed comparison governed by explicit criteria mitigates leading-question bias | Yes |
| Validated prompts automatically transfer to new model versions without regression testing | No |
Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.
Zero-shot is a solid baseline for closed-set classification and few-shot examples steer structured extraction. Chain-of-thought adds token and latency overhead so it should be applied selectively, neutral framing prevents bias, and model migrations always require regression evaluation.