← Dictionary / Zero-shot learning

Zero-shot learning

Appears in 2 papers

Performing a task without any task-specific examples or fine-tuning.

As used in Paper 10 — Improving Language Understanding by Generative Pre-Training →

Performing a task without any task-specific examples or fine-tuning. GPT-1 did not show zero-shot performance; GPT-2 and GPT-3 progressively improved in this area.

As used in Paper 12 — Language Models are Few-Shot Learners →

Performing a task with no examples in the prompt—only a description. For example, "Translate this to Hindi: Good morning" with no prior translation examples shown. Weak but enables new tasks immediately.

Appears in papers

Paper 10 — Improving Language Understanding by Generative Pre-Training → Paper 12 — Language Models are Few-Shot Learners →