Domain 3 of 4

Techniques to Improve Gen AI Model Output

The limits a foundation model arrives with, grounding against first party, third party and world data, prompting from zero shot through ReAct, the sampling parameters that shape a response, and customisation.

Concepts

~20%

Of the exam

Practice questions

Concepts in this domain

01What generative AI is good and bad atThe capabilities worth building on, the failure modes that are properties of the technology rather than bugs, and how to tell a suitable problem from an unsuitable one.02Grounding and retrievalWhat grounding means, the difference between first party, third party and world data, and the sampling parameters that shape what comes back.03Prompt engineeringThe parts a prompt is made of, the shot based techniques and chain of thought, and the four attacks the syllabus expects you to name.04Customising a foundation modelThe four ways to make a general model fit your problem, ordered by cost, and the question that tells you which one you need.

Try a question from this domain

An assistant must answer questions about legislation that changes several times a year, using current public information. Which grounding approach fits?

APrebuilt RAG with Agent Search over an internal copy of the legislation.
BFine tuning the model on the legislation each time it changes.
CRaising the context window so more of the legislation fits in each prompt.
DGrounding with Google Search, which ties answers to current public information.

12 questions on this domain.

One per page, with a worked explanation.

Start the set