AI Ultimate AI Engineer 3

Question 1

In LoRA (Low-Rank Adaptation) fine-tuning, what is decomposed into low-rank matrices?

Accepted Answer

The weight update matrices ΔW

Answer

LoRA freezes original weights and approximates ΔW = A·B where A and B are low-rank, drastically reducing trainable parameters.

Question 2

Which of the following is an example of few-shot prompting?

Accepted Answer

Including 3 example input-output pairs in the prompt before the actual query

Answer

Few-shot prompting embeds a small number of demonstration examples directly in the prompt to guide the model's output format and style.

Question 3

A production LLM application has a P99 latency of 8 seconds. Which optimization most directly addresses time-to-first-token?

Accepted Answer

Applying KV-cache to the prefill step

Answer

Caching the KV states during prefill avoids recomputing them for the prompt, directly reducing time-to-first-token latency.

Question 4

What problem does RLHF (Reinforcement Learning from Human Feedback) primarily solve?

Accepted Answer

Aligning model outputs with human preferences and values

Answer

RLHF trains a reward model on human preference data and uses RL to fine-tune the LLM to maximize that reward, improving alignment.

Question 5

Which evaluation approach detects when a model performs well on a benchmark due to memorizing test data rather than generalizing?

Accepted Answer

Contamination analysis checking if test data appears in pretraining

Answer

Contamination analysis searches the pretraining corpus for n-gram overlaps with benchmark test sets to detect data leakage.

AI - Engineer Practice Test