Question 1

When customizing an LLM for a specific industry task where quick deployment and computational efficiency are priorities, which of the following methods would be most appropriate?

Options :

A : Using the original LLM without any modifications to handle industry-specific tasks.

B : Fine-tuning an existing large model on a small, targeted dataset specific to the industry.

C : Training a new model from scratch using industry-specific data.

D : Applying data augmentation techniques on the industry-specific dataset without fine-tuning the model.

Answer: B

Question 2

In a scenario where an AI system is designed to automate document processing by classifying, summarizing, and extracting key information from large volumes of legal documents, which combination of techniques leveraging pretrained modern LLMs would be most effective?

Options :

A : Utilize a pretrained LLM solely for text classification, with separate models for summarization and key information extraction.

B : Use a rule-based system for text classification, followed by a simple RNN for summarization, and a pretrained LLM for question-answering.

C : Apply a CNN for text classification, a transformer model for summarization, and a simple linear model for question-answering.

D : Use a pretrained LLM for token classification, followed by text summarization, and then apply question-answering for key information extraction.

Answer: D

Question 3

You are designing a multimodal AI system for a smart city project that monitors traffic, air quality, and public safety using video feeds, sensor data, and social media analysis. The system needs to operate continuously with high accuracy while minimizing energy consumption across multiple edge devices. Which strategy should you prioritize to achieve the best balance between energy efficiency and system performance?

Options :

A : Implement model pruning to reduce the size of the neural networks used in the system.

B : Increase the frame rate of video feeds to capture more detailed data.

C : Deploy separate models for each data modality (video, sensor, social media) without integration.

D : Use high-resolution sensors to improve the accuracy of air quality measurements.

Answer: A

Question 4

You are working with an AI development team to integrate a multimodal generative AI model into an existing application. The model needs to generate both text and images based on user inputs. However, during the testing phase, the team encounters an issue where the image generation quality significantly degrades when the text prompts are complex. How should the team address this issue?

Options :

A : Simplify the text prompts

B : Reduce the size of the model to increase speed

C : Switch to a text-only generative model

D : Fine-tune the model on a more diverse dataset

Answer: D

Question 5

You are tasked with improving a multimodal neural network used for predicting patient outcomes based on medical imaging, lab results, and clinical notes. The current model struggles with learning complex features due to the depth of the network. Which benefit of residual connections would most directly address this problem?

Options :

A : Residual connections enable the network to learn complex features without degradation of performance

B : Residual connections allow for sequential processing of each modality independently

C : Residual connections improve the interpretability of the model’s outputs

D : Residual connections reduce the number of parameters in the network

Answer: A

Higher Test Marks with Free Online NCA-GENM Exam Practice

Buying Options: