wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

Generative AI and Transformers Quiz

Total questions: 25

Worksheet time: 13mins

Name
Class
Date
1.

What best defines Generative AI?

a)

AI that only classifies data

b)

AI models that generate new, coherent, context-aware content

c)

Rule-based automation systems

d)

AI for numerical optimization

2.

Which component enables attention-based contextual understanding?

a)

CNN

b)

RNN

c)

Transformer

d)

Decision Tree

3.

What is the purpose of foundation models?

a)

Task-specific training only

b)

Training on massive, diverse datasets

c)

Real-time data retrieval

d)

Model evaluation

4.

Which stage produces responses to user queries?

a)

Pre-training

b)

Fine-tuning

c)

Inference

d)

Feedback

5.

Why is ethical and responsible AI important in GenAI?

a)

To reduce training cost

b)

To avoid plagiarism, bias, and misuse

c)

To increase creativity

d)

To speed up inference

6.

What key limitation of RNNs did Transformers solve?

a)

Overfitting

b)

Sequential processing speed

c)

Vocabulary size

d)

Tokenization issues

7.

Why do Transformers support faster training?

a)

Smaller models

b)

Parallel token processing

c)

No attention layers

d)

Fewer parameters

8.

Which are the two core blocks of a Transformer?

a)

Input & Output

b)

Encoder & Decoder

c)

CNN & RNN

d)

Tokenizer & Classifier

9.

What is the primary role of the encoder?

a)

Generate next tokens

b)

Convert input into contextual representations

c)

Mask future tokens

d)

Apply softmax

10.

Why is positional encoding required?

a)

To reduce parameters

b)

Transformers lack inherent word-order awareness

c)

To remove attention

d)

To improve embeddings size

11.

What is the benefit of multi-head attention?

a)

Reduces memory usage

b)

Captures multiple relationships in parallel

c)

Removes feed-forward layers

d)

Eliminates positional encoding

12.

What best describes a Large Language Model (LLM)?

a)

Rule-based NLP system

b)

Deep neural network trained on massive text data

c)

Database query engine

d)

Symbolic reasoning system

13.

What enabled the rise of LLMs?

a)

Smaller datasets

b)

Manual feature engineering

c)

Transformers + massive data + compute power

d)

Keyword matching

14.

Which pre-training objective is used by BERT?

a)

Next-word prediction

b)

Masked language modeling

c)

Reinforcement learning

d)

Text-to-text learning

15.

GPT models are primarily designed for:

a)

Bidirectional understanding

b)

Step-by-step text generation

c)

Token masking

d)

Classification tasks only

16.

What makes T5 unique?

a)

Decoder-only architecture

b)

Frames all NLP tasks as text-to-text

c)

Vision-only model

d)

No fine-tuning required

17.

IBM Granite models are mainly built for:

a)

Entertainment chatbots

b)

Academic research only

c)

Enterprise-grade AI workflows

d)

Gaming applications

18.

A key focus of Granite models is:

a)

Creativity over accuracy

b)

Trust, transparency, and reliability

c)

Image generation

d)

Social media content

19.

Why are Granite models suitable for enterprise use?

a)

They are closed-source

b)

They emphasize governed, factual outputs

c)

They avoid fine-tuning

d)

They remove evaluation

20.

What primary problem in LLMs does RAG address?

a)

Slow inference

b)

Hallucination and outdated information

c)

Token limits

d)

Poor embeddings

21.

What happens during Context Injection in RAG?

a)

Model training

b)

Retrieved documents added to the prompt

c)

Tokenization

d)

Model evaluation

22.

Why does RAG reduce retraining needs?

a)

It updates model weights automatically

b)

It changes data instead of the model

c)

It removes embeddings

d)

It avoids inference

23.

In a text summarization project, the primary goal is to:

a)

Translate text into another language

b)

Generate a shorter version while preserving key meaning

c)

Classify documents into labels

d)

Detect grammatical errors

24.

Which task best represents a text classification project?

a)

Generating code from prompts

b)

Assigning labels like sentiment or topic to text

c)

Fixing syntax errors in code

d)

Summarizing research papers

25.

In code generation and bug-fixing projects, the model is expected to:

a)

Only explain code

b)

Generate, analyze, and correct faulty code logic

c)

Compile programs

d)

Replace developers