AIF-C01 Question 75

Community Discussion

Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.

• Chose C4

  C. Pairs of user messages and correct user intents: Few-shot learning works by providing the model with a small number of labeled examples to help it learn how to generalize better. In this case, the model needs to be trained with examples that consist of user messages and their corresponding intents. These pairs will help the LLM improve its ability to classify new user messages into the correct intent categories. The model will use these few-shot examples to adjust its response pattern to better detect the user's intent.

• Chose C1

  C. Pairs of user messages and correct user intents Few-shot learning is a machine learning technique that allows the model to learn from small amounts of data, including labeled examples or "shots." In this case, the company wants to use few-shot learning to improve intent detection accuracy. To implement few-shot learning for intent detection, the company needs additional data in the form of pairs of user messages and their corresponding correct user intents. This data will serve as the "shooting" examples that the LLM can learn from.

• Chose C1

  C. Pairs of user messages and correct user intents Explanation: Few-shot learning involves training a model with a small number of examples (or samples). In this case, the goal is to improve intent detection, which requires a clear understanding of the user's intent based on their message. To fine-tune the large language model (LLM) using few-shot learning, the model needs examples of user messages along with their corresponding correct user intents. These pairs will teach the model how to accurately classify user intents based on input messages.