Zero Shot vs One Shot

In Machine learning terms:

Zero-shot learning is a machine learning technique that allows models to recognize objects they have never seen before. It works by training models on a set of known objects and then using that knowledge to recognize new objects that share similar attributes 1. For example, if a model is trained on images of dogs and cats, it can recognize images of other animals like rabbits or squirrels, even though it has never seen those animals before.

One-shot learning, on the other hand, is a technique that allows models to recognize objects based on a single example. It works by training models on a set of known objects and then using that knowledge to recognize new objects based on a single example 2. For example, if a model is trained on images of dogs and cats, it can recognize a new image of a dog or cat based on a single example.

In summary, zero-shot learning is used to recognize objects that are similar to known objects, while one-shot learning is used to recognize objects based on a single example. Both techniques have their own strengths and weaknesses, and the choice of which technique to use depends on the specific problem you are trying to solve.

With regards to Prompting or utilizing the LLMs or Copilots

Zero Shot or One Shot Prompting

Give a few concrete examples

- Zero shot prompting is when a model is given a task without any prior examples or instructions. For example, asking a model to write a summary of a text without showing it any summaries before.

- One shot prompting is when a model is given a task with one example or instruction. For example, asking a model to write a summary of a text after showing it one summary of another text.

I hope this information helps. If you have any other questions or need further assistance, please let me know. ??