AI AI Engineer: NLP and Large Language Models 3

Question 1

What is 'vector similarity search' used for in LLM-powered applications?

Accepted Answer

Finding the most semantically similar documents to a query by comparing embedding vectors

Answer

Training transformer attention weights

Answer

Reducing LLM inference latency

Answer

Detecting prompt injection attacks

Question 2

Which metric measures how well an LLM predicts a test corpus, with lower values indicating better language modeling?

Accepted Answer

Perplexity

Answer

BLEU

Answer

Precision

Answer

AUC-ROC

Question 3

What is 'chain-of-thought prompting' in LLMs?

Accepted Answer

Prompting the model to reason through intermediate steps before giving a final answer

Answer

Chaining multiple LLM API calls sequentially

Answer

Training the model on logical reasoning datasets

Answer

Using multiple models in a pipeline

Question 4

What is a vector database's primary role in an LLM application architecture?

Accepted Answer

Storing and querying high-dimensional embeddings for semantic similarity search

Answer

Storing model weights for fast loading

Answer

Caching LLM API responses

Answer

Managing prompt templates

Question 5

What is 'semantic chunking' in the context of building RAG systems?

Accepted Answer

Dividing documents into chunks based on semantic coherence to preserve meaningful context

Answer

Splitting documents by fixed token counts

Answer

Encrypting document chunks before embedding

Answer

Filtering out low-quality document segments

Question 6

What does 'temperature' control in LLM text generation?

Accepted Answer

The randomness of token sampling — higher values produce more diverse outputs

Answer

The computational load during inference

Answer

The maximum length of generated text

Answer

The model's confidence threshold for answering