Interactive and Visual Prompt Engineering for Ad-hoc Task Adaptation with Large Language Models

TL;DR

This paper introduces PromptIDE, a visual tool that enables users to experiment with, optimize, and deploy prompts for large language models to improve performance on new, ad-hoc NLP tasks.

Contribution

The paper presents a novel interactive workflow and visualization tool for prompt engineering, facilitating efficient prompt optimization for large language models.

Findings

PromptIDE helps users identify effective prompts through visualization and iterative testing.

The workflow supports transitioning from small data feedback to large data validation.

Demonstrations show improved task performance with optimized prompts.

Abstract

State-of-the-art neural language models can now be used to solve ad-hoc language tasks through zero-shot prompting without the need for supervised training. This approach has gained popularity in recent years, and researchers have demonstrated prompts that achieve strong accuracy on specific NLP tasks. However, finding a prompt for new tasks requires experimentation. Different prompt templates with different wording choices lead to significant accuracy differences. PromptIDE allows users to experiment with prompt variations, visualize prompt performance, and iteratively optimize prompts. We developed a workflow that allows users to first focus on model feedback using small data before moving on to a large data regime that allows empirical grounding of promising prompts using quantitative measures of the task. The tool then allows easy deployment of the newly created ad-hoc models. We demonstrate the utility of PromptIDE (demo at http://prompt.vizhub.ai) and our workflow using several real-world use cases.