AI-900 Practice Question: Describe features of generative AI workloads on Azure
What is fine-tuning in the context of large language models?
⚠ Common exam trap
It's easy for candidates to confuse fine-tuning with inference optimization or model compression, because all three can improve performance in production, but only fine-tuning actually modifies model weights through additional training on domain-specific data.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Training a pre-trained model further on domain-specific data to improve task performance
Fine-tuning takes a pre-trained large language model (LLM) and continues the training process on a smaller, domain-specific dataset. This adjusts the model's weights to specialize its outputs for particular tasks (e.g., legal document summarization or medical Q&A) without retraining from scratch. It is distinct from prompt engineering or retrieval-augmented generation because it permanently modifies the model parameters.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Adjusting the model's response speed for production deployment
Why it's wrong here
Adjusting the model's response speed for production deployment is an inference-optimization concern—using hardware acceleration, model batching, or runtime serving choices—and does not involve any supervised or unsupervised training. Unlike fine-tuning, which updates model weights during a separate training phase based on domain-labeled data, response-speed tuning occurs after training and does not change the model's learned capabilities or task accuracy. Confusing deployment latency tuning with fine-tuning conflates operational infrastructure with model adaptation.
- ✓
Training a pre-trained model further on domain-specific data to improve task performance
Why this is correct
Fine-tuning takes a foundation model that was pre-trained on massive, general corpora and performs additional supervised training on a smaller, labeled dataset from the target domain, updating all or some weights through backpropagation. This process lets the model leverage its prior linguistic or visual knowledge and adjust its internal representations to the vocabulary, style, and label distribution of that domain, often producing large accuracy gains while requiring much less data and compute than training from scratch. In Azure AI, this is the core adaptation step for models such as GPT, BERT, or ResNet variants when customizing them for industry-specific tasks.
- ✗
Manually reviewing and correcting model outputs
Why it's wrong here
Manually reviewing and correcting model outputs is a human-in-the-loop quality-control step, often used for audit or immediate error patching, but it does not modify the trained model's underlying weights. Fine-tuning, in contrast, is an automated training procedure that updates the model's parameters by backpropagating loss gradients across a labeled dataset for the target task. Reviewing outputs is essentially inference-time feedback and evaluation, while fine-tuning is a data-driven, training-time adaptation.
- ✗
Compressing a large model into a smaller, faster version
Why it's wrong here
Compressing a large model into a smaller, faster version refers to techniques such as knowledge distillation, pruning, or quantization that reduce memory footprint and inference latency, often with some accuracy trade-off. These compression methods change the model's representation or numerical precision but do not retrain the model on new task-specific data. Fine-tuning is fundamentally different: it starts from a pre-trained model and continues gradient-based training on domain-specific examples to alter its behavior toward a particular task.
Go deeper
Related to this question
Learn chapter
Azure Document Intelligence
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
Key term
Prompt engineering
Prompt engineering is the practice of designing and refining input queries to AI models to get the most accurate, relevant, and useful outputs.
About these practice questions
This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.