Prompting Workshop

Front

What is the main purpose of prompting in LLMs?

Back

To guide the model's response without additional training

Front

What is Fine-Tuning in the context of LLMs?

Back

The process of adapting a pre-trained LLM to a specific task using additional training data

Front

Which of the following is NOT a limitation of prompting? Sensitivity to prompt design, Inability to handle misinformation, The need for structured reasoning in complex tasks, The requirement to modify the model’s architecture

Back

The requirement to modify the model’s architecture

Front

What is the difference between zero-shot and few-shot prompting?

Back

Zero-shot provides no examples, while few-shot provides a few examples

Front

How does Chain-of-Thought (CoT) prompting improve model responses?

Back

By allowing the model to explain its reasoning step-by-step

Front

Which of the following is an example of Zero-Shot Chain-of-Thought Prompting? Options: "What is the capital of France?", "Think step-by-step: Who is taller, a giraffe or a dog?", "Translate 'Hello' to French.", "Give me a summary of this article."

Back

"Think step-by-step: Who is taller, a giraffe or a dog?"

Front

What is one major benefit of in-context learning?

Back

It allows LLMs to learn new tasks without weight updates

Front

What is the primary factor limiting the effectiveness of prompting?

Back

The quality and structure of the prompt

Front

Which of the following is true about preference alignment in LLMs? It is achieved through Reinforcement Learning with Human Feedback (RLHF), It eliminates all biases in model outputs, It involves increasing the dataset size exponentially, It removes the need for any prompting

Back

It is achieved through Reinforcement Learning with Human Feedback (RLHF)

Front

Why do induction heads in LLMs help with in-context learning?

Back

They allow the model to recognize and complete patterns from input sequences