Prompt2Model: Generating Deployable Models from Natural Language Instructions

TL;DR

Prompt2Model converts natural language task descriptions into specialized, efficient models through retrieval, dataset generation, and fine-tuning, outperforming large language models like GPT-3.5-turbo while being significantly smaller.

Contribution

It introduces a novel method to generate deployable models from natural language prompts, combining retrieval, dataset creation, and supervised fine-tuning, improving efficiency and performance.

Findings

Models trained with Prompt2Model outperform GPT-3.5-turbo by 20% on three tasks.

Generated models are up to 700 times smaller than large language models.

The approach enables reliable performance estimation before deployment.

Abstract

Large language models (LLMs) enable system builders today to create competent NLP systems through prompting, where they only need to describe the task in natural language and provide a few examples. However, in other ways, LLMs are a step backward from traditional special-purpose NLP models; they require extensive computational resources for deployment and can be gated behind APIs. In this paper, we propose Prompt2Model, a general-purpose method that takes a natural language task description like the prompts provided to LLMs, and uses it to train a special-purpose model that is conducive to deployment. This is done through a multi-step process of retrieval of existing datasets and pretrained models, dataset generation using LLMs, and supervised fine-tuning on these retrieved and generated datasets. Over three tasks, we demonstrate that given the same few-shot prompt as input, Prompt2Model trains models that outperform the results of a strong LLM, gpt-3.5-turbo, by an average of 20% while being up to 700 times smaller. We also show that this data can be used to obtain reliable performance estimates of model performance, enabling model developers to assess model reliability before deployment. Prompt2Model is available open-source at https://github.com/neulab/prompt2model.