---
date: '2024-02-12'
description: instruction-tuned language models performing tasks without examples, outperforming larger models through improved zero-shot generalization.
id: zero-shot learning
modified: 2026-06-05 15:08:29 GMT-04:00
tags:
  - llm
title: zero-shot prompting
created: '2024-02-12'
published: '2024-02-12'
pageLayout: default
slug: thoughts/zero-shot-learning
permalink: https://aarnphm.xyz/thoughts/zero-shot-learning.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
[Finetuned Language Models Are Zero-Shot Learners](https://arxiv.org/abs/2109.01652.pdf) \[@wei2022finetunedlanguagemodelszeroshot\] 

The paper argues that zero-shot prompting on a instruction-tuned small language models outperform [[thoughts/LLMs]] systems.

- Instruction-tuning actually improve zero-shot learning performance.
- Mostly tested on FLAN, but show results throughout with GPT-3 and on few reading comprehension dataset.

Honorable mentions include prompt tuning or few-shots prompting