The correct answer is: **2. All of the listed options.** All of the options provided (non-representative training data, corrupted data, and insufficient training data) can lead to sub-optimal generative AI models.

The correct answer is **3. All of the listed options.** Generative AI can be used for new content creation, AI for customer interaction, and powering chatbots, among other applications.

1. Write clear and specific prompts.

The biggest strength of LLMs (Large Language Models) is that **1. They were trained on extremely large amounts of text.** This extensive training enables them to understand and generate human-like language, respond to queries, and grasp a wide range of topics.

Generative AI models mostly use **2. Artificial Neural Networks**. Specifically, architectures like Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) are examples of generative models founded on neural networks.

The tool used for programming code generation among the options provided is **3. GitHub Copilot**.

The best practice for storing an OpenAI API key among the options you've provided would be **2. Environment variable**. Using environment variables helps keep sensitive information like API keys secure and out of your source code, reducing the risk of accidental exposure in version control systems.

Diffusion models work by gradually adding noise to an image in a controlled manner and then learning to reverse this process to generate new images. Thus, the correct option among the ones you provided is: **2. They gradually add noise into the original image.** In detail, diffusion models start with an original image and iteratively add Gaussian noise over several time steps. During training, the model learns to predict and remove this noise

The best option that relates to the integration of ChatGPT with other systems is **3. API Key**. ChatGPT can be integrated into various applications and systems using an API, which requires an API key for authentication. This allows developers to incorporate ChatGPT's capabilities into their own platforms, such as chatbots, customer support systems, and other applications, leveraging the language model's functionalities.

3. All of the listed options. Improving the content created in response to a prompt can be achieved through a combination of writing clear instructions, writing specific instructions, and allowing extra time for the model to think. Each of these strategies contributes to enhancing the quality and relevance of the generated content. Clear and specific instructions help guide the model effectively, and giving it more time can lead to more

A typical challenge of Generative AI is **2. Difficult to create high quality content**. While generative models can produce a wide range of outputs, ensuring that these outputs are not only coherent but also of high quality and relevant to a specific context can be quite challenging. The models may struggle with issues such as maintaining consistency, avoiding biases, and generating content that meets human standards of creativity and relevance.

The correct option that defines self-attention mechanism is: 4. Representation of a sentence is computed by relating different words in it. Self-attention allows a model to weigh the significance of different words in a sentence when generating a representation, enabling it to capture contextual relationships effectively.

Let's address each question one by one: ### Question 23 **Which one of the following is not a staple food crop?** - a. Bread - b. Maize - c. Rice - d. Yams **Answer:** **a. Bread** - Bread is a food product made from staple crops like wheat, but it is not a staple food crop itself. --- ### Question 24 **Which one of the following observations from the accompanying figure is true?** - a. Solar was the most energy-demanding renewable source

1. System.

To evaluate the statements based on the unspecified figure on global primary energy consumption by source, we can analyze each statement considering historical trends and general knowledge of energy consumption: a. **Solar was the most energy-demanding renewable source in 1970.** - This statement is likely false, as solar energy consumption was negligible in 1970 compared to other renewable sources like hydropower. b. **The difference in

The correct option is: **3. Instruction-tuned LLMs are for more refined results.** Instruction-tuned LLMs are specifically trained to follow instructions and produce outputs that align more closely with user expectations, making them generally more refined in their responses compared to base LLMs. Base LLMs are typically more general and lack this specific tuning for following instructions.

a. Bread Bread is a food product made from staple food crops like wheat, but it is not itself a staple food crop. The other options (maize, rice, yams) are all considered staple food crops.

The key component in Transformer models is **2. Self-Attention mechanism**.