問題文
A tester holds a screenshot of a checkout page and the user story the page was built from. The tester wants a model to point out the places where the screen and the wording of the story disagree. Which kind of model does the syllabus present for that request?
選択肢
- Vision-language AI models, because they contrast the screen capture with the written specification and indicate the visual discrepancies they find.
- Single-mode chat assistants, because they compare the token count with the written specification and report the shorter sentences they find.
- Rule-based symbolic engines, because they compare the stored table with the written specification and report the missing rows they find.
- Keyword-driven test frameworks, because they compare the stored keyword with the written specification and report the unmapped operations they find.