Basics of Generative AI and LLMs CT-GENAI Practice Questions (Page 5)

Questions 21–25

1. 21foundation · easy

   Which of the following is a main capability of Large Language Models?

   • AProviding real-time weather forecasts based on satellite data.
   • BExecuting code and managing cloud infrastructure.
   • CPerforming high-precision arithmetic calculations.
   • DGenerating human-like text, summarizing documents, translating languages, and answering questions.
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A test automation engineer wants to use AI to generate new test data for a login form, including valid, invalid, and boundary-case usernames and passwords. Which type of AI approach is best suited for this task?

   • AA discriminative AI model that classifies existing login attempts as valid or invalid
   • BA generative AI model that learns patterns from existing user data and produces new, plausible data points
   • CA rule-based script that randomly combines characters to produce usernames and passwords
   • DA supervised learning model trained to predict whether a given password is strong or weak
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   In software testing, which task is an example of using generative AI rather than traditional (discriminative) AI?

   • AClustering similar bug reports to identify common root causes.
   • BPredicting which software modules are most likely to contain defects.
   • CGenerating realistic synthetic test data for a new user registration form.
   • DClassifying customer feedback as positive or negative sentiment.
   Reveal Solution Discussion Select an answer first
5. 24application · medium

   A test data engineer wants to generate synthetic user profiles for performance testing. The profiles must resemble real user behavior patterns. What is the most effective way to use generative AI for this task?

   • AUse a rule-based system to combine predefined attributes into profiles
   • BTrain a generative model on a dataset of real user profiles and use it to produce new, realistic profiles
   • CUse a pre-trained LLM to generate random profiles without any training on user data
   • DUse a discriminative model to classify existing profiles and select the most representative ones
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   What is a key characteristic of Large Language Models (LLMs) like GPT?

   • AThey are trained on vast amounts of text data and have billions of parameters.
   • BThey require a human to manually encode every grammar rule.
   • CThey are designed to process only numerical data, not text.
   • DThey are limited to translating between two specific languages.
   Reveal Solution Discussion Select an answer first