Google

A model's ability to understand instructions, an emergent capability from pre-training, is a key factor that enables zero-shot learning. This allows the model to generalize its knowledge to perform new tasks for which it has not been explicitly trained, simply by following novel textual instructions it has never encountered before.

Enabling Zero-Shot Learning through Instruction Understanding

Given the following scenario, what would be the most logical and cost-effective *first step* for the startup to validate their new feature idea? Justify your recommendation by explaining the underlying capability of the model that makes this step viable.

LLM Feature Development Strategy

A colleague argues that a large language model's ability to perform a completely new task (e.g., 'Translate this English sentence into pirate slang') without specific training is simply because it has memorized a vast number of similar translation examples. Evaluate this argument. In your response, explain the more fundamental capability that enables this performance and how it is developed.

Evaluating the Source of Zero-Shot Capabilities

A large language model, pre-trained on a diverse corpus of internet text, is given the following prompt for the first time: "Rephrase the following sentence in the style of a 1920s radio announcer: 'The new smartphone will be available next week.'" The model successfully responds with something like: "Hark, good citizens! The magnificent new pocket telephone contraption shall grace our stores in but a week's time!" Which statement best analyzes the primary reason the model can successfully compl

Learn Before

Related