Free AWS MLA-C01 Actual Exam Questions - Question 10 Discussion
An ML engineer is building a generative AI application on Amazon Bedrock by using large language
models (LLMs).
Select the correct generative AI term from the following list for each description. Each term should
be selected one time or not at all. (Select three.)
• Embedding
• Retrieval Augmented Generation (RAG)
• Temperature
• Token
Embedding, RAG, and Temperature fit best since Token is basic input unit.
Embedding, RAG, and Temperature cover more specific generative AI roles than Token.
Token is just the smallest text piece, so it’s pretty basic. Embedding and RAG definitely fit since they involve data prep and retrieval. Temperature is a good pick too because it controls randomness in output generation.
Token seems more about text units, which feels basic rather than a unique generative AI concept here. So picking Embedding, RAG, and Temperature makes sense since they cover vectorizing, info retrieval, and output variety.
I think Embedding fits because it’s about turning text into vectors, which is key before feeding info into the model. RAG makes sense since it’s about pulling in external knowledge to improve responses. Temperature definitely relates to how creative or random the generated text is, so it matches the control of output style. Token feels more like the basic unit of text but not a unique generative AI concept like the others here. So I’d go with B, D, and C based on function and relevance to generation and retrieval.
Embedding clearly matches vector representation, while RAG is about external data use.
Temperature controls randomness, so that’s the one for tweaking output creativity.
Tokens are just chunks of text, so that aligns with the splitting description.
For the part about breaking down text into pieces, that’s definitely Token. You can eliminate Temperature for that since it deals with randomness, not text handling. So Embedding fits the vector representation, RAG is about using external info to boost responses, and Token is the smallest unit of text. That covers the three terms neatly without overlap.
For the description about converting text into vectors, the correct term is Embedding. The part about improving answers using external data fits Retrieval Augmented Generation (RAG). And the one about controlling randomness matches Temperature.