Zero-Shot Learning
Definition
The ability of a model to perform a task it has never been explicitly trained on, without any task-specific examples provided.
Zero-shot learning represents one of the most remarkable capabilities of large AI models. A model performs a task based solely on its general knowledge and the task description, without seeing any examples. For instance, a large language model can translate between languages it wasn't specifically trained to translate by understanding the concept of translation from its training data. Zero-shot capabilities scale with model size and the diversity of pre-training data. GPT-3 demonstrated strong zero-shot performance across many tasks, and subsequent models have continued to improve. Zero-shot learning is a key measure of how well a model generalizes.
Related Terms
Transfer Learning
A technique where a model trained on one task is reused or adapted for a different but related task,...
Few-Shot Learning
The ability of an AI model to learn and generalize from only a handful of examples, rather than requ...
Foundation Model
A large AI model trained on broad data at scale that can be adapted to a wide range of downstream ta...
Large Language Model
A neural network with billions of parameters trained on massive text datasets, capable of understand...