
Answer-first summary for fast verification
Answer: Negative Prompting
## Explanation **Negative Prompting** is a technique used in generative AI where users specify what they **do not want** to appear in the generated output. In this case: - The user wants a "Thanksgiving dinner scene" (positive prompt) - The user also specifies "Do not include any people or brand logos in the image" (negative prompt) **Why other options are incorrect:** - **A) Few-shot Prompting**: This involves providing examples (shots) to guide the model's output, which is not happening here. - **C) Reinforcement Learning**: This is a training method where models learn through rewards/penalties, not a prompting technique. - **D) Contextual Embedding**: This refers to how words/sentences are represented in vector space based on their context, not a prompting technique. **Key characteristics of Negative Prompting:** 1. Explicitly states what should be excluded from the output 2. Helps refine and control the generated content 3. Particularly useful in image generation to avoid unwanted elements 4. Works by guiding the model away from certain concepts during the generation process
Author: Jin H
Ultimate access to all questions.
No comments yet.
A generative image model is asked to create a "Thanksgiving dinner scene" but the user specifies: "Do not include any people or brand logos in the image." Which technique is being used?
A
Few-shot Prompting
B
Negative Prompting
C
Reinforcement Learning
D
Contextual Embedding