0% found this document useful (0 votes)
14 views

set2 (1)

The document outlines an exam for a course on Generative AI, consisting of multiple-choice questions that assess understanding of various prompt patterns and their applications in large language models. Each question carries a specific mark value, with penalties for incorrect answers, and emphasizes the importance of user interaction and contextual information in AI models. The scenarios provided in the questions illustrate practical applications of generative AI in language learning, personalized recommendations, and prompt engineering.

Uploaded by

Athul
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
14 views

set2 (1)

The document outlines an exam for a course on Generative AI, consisting of multiple-choice questions that assess understanding of various prompt patterns and their applications in large language models. Each question carries a specific mark value, with penalties for incorrect answers, and emphasizes the importance of user interaction and contextual information in AI models. The scenarios provided in the questions illustrate practical applications of generative AI in language learning, personalized recommendations, and prompt engineering.

Uploaded by

Athul
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

INT426 (Generative AI)

Section: CA-2 Roll No:

Max Marks: 30 Duration: 40 mins

Instructions
1. Each questions carries 2 mark, wrong answer will detect 0.5 marks.
2. Overwriting of answer will not be considered.
3. Choose the correct answer and write in the cell given below

Q1 Q6 Q11
Q2 Q7 Q12
Q3 Q8 Q13
Q4 Q9 Q14
Q5 Q10 Q15
1.How does the question refinement pattern pattern tailors the output based on the intended
contribute to the user experience when interacting audience's characteristics.
with large language models? c) The persona pattern restricts the types of questions
a) By limiting the user's ability to ask questions freely that can be asked, while the audience persona pattern
b) By encouraging users to provide more context in restricts the types of responses that can be generated. d)
their queries The persona pattern automates the process of
c) By automating the process of generating questions generating user-specific content, while the audience
without user input persona pattern automates the process of generating
d) By decreasing the responsiveness of the language audience-specific content.
model to user queries 4. How does the flipped interaction pattern differ
2. What is the purpose of the cognitive verifier from traditional interaction with large language
pattern described ? models?
a) It places the user in a passive role, answering
a) To verify the accuracy of information provided by
questions posed by the language model.
the language model. b) It encourages users to dictate the specific outputs
b) To encourage users to think critically about their generated by the language model.
questions. c) It facilitates collaborative problem-solving by
c) To generate additional questions that aid in allowing users to provide guidance to the language
problem-solving. model.
d) To limit the scope of questions asked to the d) It restricts the language model's ability to ask
questions and obtain information from the user.
language model.
5. During prompt execution in template generation,
3.How does the audience persona pattern differ
which component is commonly substituted?
from the persona pattern in interacting with large
a. Placeholders
language models?
b. Keywords
a) The persona pattern focuses on the identity of the
c. Variables
user, while the audience persona pattern focuses on the
d. Punctuations
identity of the audience.
6. How does the approach of few-shot prompting
b) The persona pattern tailors the output based on the
diverge from traditional methods in training large
user's characteristics, while the audience persona
language models?
a) It hinges on furnishing instances of inputs and a. Automatic speech recognition
anticipated outcomes. b. Voice command recognition
b) It engrosses the model in generating responses c. Speaker diarization
devoid of external input. d. Audio segmentation
c) It depends on explicit rules and guidelines for task 12. How does incorporating specific words like
execution. "title" and "author" in a prompt influence the
d) It concentrates on circumscribing the model's output of large language models?
autonomy in output generation.
A) It guides the model towards generating output in a
7. How does the inclusion of contextual information structured format
impact the effectiveness of few-shot examples? B) It introduces variability in model responses
a) It confuses the model by providing irrelevant details C) It restricts the model's creativity
b) It enhances the model's ability to derive underlying D) It has no effect on the model's output
processes 13.) Scenario: You are designing prompts to guide
c) It limits the model's flexibility in generating diverse users in learning a new language through
outputs vocabulary exercises. Which prompt pattern would
d) It reduces the model's reliance on prior knowledge be most effective in this context?
for decision-making A) The Game play Pattern
8. Scenario: You are designing a language learning B) The Problem-Solution Pattern
app that offers personalized exercises based on C) The Scaffolded Learning Pattern
users' proficiency levels and learning preferences. D) The Storytelling Pattern
Which prompt pattern would be most suitable for 14) Scenario: You are creating prompts to teach
users how to bake a classic chocolate chip cookie
creating dynamic and adaptive learning prompts?
from scratch. Which prompt pattern would be most
A) The Fact check Pattern aligned with this objective?
B) The Combining Pattern A) The Argumentation Pattern
C) The Review Pattern B) The Context Manager Pattern
D) The Context Manager Pattern C) The Recipe Pattern
9. Which of the following best describes the purpose D) The Review Pattern
of chain of thought prompting in training large 15) Scenario: You are creating a virtual assistant
application designed to provide users with
language models?
personalized recommendations for leisure activities
a) To challenge the model with ambiguous tasks for based on their preferences and past interactions. To
better learning make the recommendations more engaging, you
b) To present questions and answers without context or decide to incorporate a feature that generates
explanation whimsical stories or scenarios related to each
recommended activity. Which prompt pattern
c) To demonstrate the step-by-step reasoning behind a
would be most appropriate for generating these tail-
solution end additions to the recommendations?
d) To structure prompts with complex vocabulary to
test comprehension A) The Tail Generation Pattern
10. . Which of the following is NOT a typical B) The Semantic Filter Pattern
application of prompt engineering in programming C) The Menu Action Pattern
assistance systems? D) The Recipe Pattern
a. Debugging
b. Software documentation
c. Code optimization
d. Test case generation
11. Which of the following is NOT a typical
application of prompt engineering in speech
recognition systems?

You might also like