Relativity Generative AI Pro Exam 2025 Questions and Answers

Relativity Generative AI Pro Exam 2025
Questions and Answers




1. 1.What are hallucinations?
zm zm zm




a. Results in which a large language model (LLM) corrects
zm zm zm zm zm zm zm zm




incongruencies in the prompt.
zm zm zm zm




b. Results in which a large language model (LLM) ignores
zm zm zm zm zm zm zm zm




questions asked in the prompt.
zm zm zm zm zm




c. Results in which a large language model (LLM) presents
zm zm zm zm zm zm zm zm




factuallyincorrectinformation.:c.Resultsinwhichalargelanguagemodel(LLM)
zm zm zm zm zm zm zm zm zm zm zm




presents
zm




factually incorrect information.
zm zm





zm zm

,2. 2.Which of the following is an advantage of generative AI
zm zm zm zm zm zm zm zm zm zm




compared with other types of machine learning AI?
zm zm zm zm zm zm zm zm




a. Generative AI does not need any additionalmodel zm zm zm zm zm zm zm




training.
zm




b. Generative AI is always faster at making predictions. zm zm zm zm zm zm zm




c. Generative AI is less computationally expensive.:a.Generative AI doesnot zm zm zm zm zm zm zm zm zm zm




need any additional model
zm zm zm zm




training.

3. 3.How is a SupportVectorMachine (SVM),the underlying AI
zm zm zm zm zm zm zm zm zm zm




model of Review Center prioritized queues, different than
zm zm zm zm zm zm zm zm




generative AI?
zm zm




a. An SVM uses example documents insteadofreview
zm zm zm zm zm zm zm




instructions to classify documents.
zm zm zm zm




b. An SVMevaluates documents independently anddoes
zm zm zm zm zm zm




not learn from previous documents.
zm zm zm zm zm




c. An SVM is pretrained on billions of documents from open
zm zm zm zm zm zm zm zm zm




databases and the web.:a.An SVMuses example documents instead of review
zm zm zm zm zm zm zm zm zm zm zm zm




instructions to classify documents.
zm zm zm zm




4. 4.What are large language models (LLMs)?
zm zm zm zm zm zm





zm zm

, a. AI models that interpret and generate responses based on
zm zm zm zm zm zm zm zm




human language.
zm zm




b. AI models that are specifically trained to generate binary
zm zm zm zm zm zm zm zm




classifications for a given task.
zm zm zm zm zm




c. AI models that can predict future trends in continuous
zm zm zm zm zm zm zm zm




data like metrics over time.:a.AI models that interpret and generate responses based
zm zm zm zm zm zm zm zm zm zm zm zm zm




on human language.
zm zm




5. 5.What does grounding refer to?
zm zm zm zm zm




a. Preventinga prompt from running because its content zm zm zm zm zm zm zm




exceeds the limits of the context window.
zm zm zm zm zm zm zm




b. The practice of requiring an AI model to
zm zm zm zm zm zm zm




provideresponsesbasedonverifiablesourceinformation to
zm zm zm zm zm zm zm




mitigate hallucinations.
zm zm




c. The basic training or instructions given to an AI model in a
zm zm zm zm zm zm zm zm zm zm zm




particular subject.:b.The practice of requiring an AI model to
zm zm zm zm zm zm zm zm zm zm zm




provide responses based on verifiable source information
zm zm zm zm zm zm zm




to mitigate hallucinations.
zm zm




6. 6.What is the primary driver behind the Relativity AI principle
zm zm zm zm zm zm zm zm zm zm




regarding AI development?
zm zm zm




a. AI development is drivenby pursuingcutting-edge
zm zm zm zm zm zm




technologies in order to lead industry trends.
zm zm zm zm zm zm zm




b. AI development is driven by prioritizing AI software
zm zm zm zm zm zm zm




capabilities.
zm





zm zm