liveBook · Manning

Zero-shot learning

As explained in LLMs in Production

A learning or inference setting in which a model performs a task without being given task-specific examples.

Zero-shot learning lets a large language model perform a completely novel task through prompting, without task-specific examples or fine-tuning. Its value is rapid specialization, but outputs can still be unintended. section 2.4

Distinguish zero-shot prompting from nearby approaches:

  • Zero-shot provides no examples; one-shot provides one example; few-shot provides several examples. section 2.4
  • Fine-tuning changes the model using labeled data, whereas prompting specializes behavior without additional training. section 2.4
  • Larger models generally show stronger zero-shot capabilities and better handling of linguistic ambiguity. section 2.4

How LLMs in Production covers it

How deep it goes
Defines zero-shot prompting and contrasts it with one-shot and few-shot prompting, but does not detail implementation procedures.
Examples
Uses specialized tasks performed through simple prompts and highlights unintended outputs from generative models.
What you can do after
Evaluate/understand when a language model can attempt a novel task without labeled examples or fine-tuning.
How it connects
Connects zero-shot learning with large language models, few-shot prompting, fine-tuning, prompt engineering, and reinforcement learning from human feedback.
Cover of LLMs in Production
liveBookmanning.com
Read LLMs in Production on liveBook Get the book at manning.com

Also explained in

Compare how 10 books explain zero-shot learning →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of LLMs in Production. The section links point to the parts of the book it draws on.