Neural Prompt Search

TL;DR

This paper introduces NOAH, a neural architecture search method for designing optimal prompt modules in vision models, improving parameter-efficient tuning across diverse datasets with strong few-shot and domain generalization capabilities.

Contribution

It proposes a novel neural architecture search approach to automatically design prompt modules for vision models, reducing manual tuning efforts.

Findings

NOAH outperforms individual prompt modules on multiple datasets.

It demonstrates strong few-shot learning ability.

It generalizes well across different domains.

Abstract

The size of vision models has grown exponentially over the last few years, especially after the emergence of Vision Transformer. This has motivated the development of parameter-efficient tuning methods, such as learning adapter layers or visual prompt tokens, which allow a tiny portion of model parameters to be trained whereas the vast majority obtained from pre-training are frozen. However, designing a proper tuning method is non-trivial: one might need to try out a lengthy list of design choices, not to mention that each downstream dataset often requires custom designs. In this paper, we view the existing parameter-efficient tuning methods as "prompt modules" and propose Neural prOmpt seArcH (NOAH), a novel approach that learns, for large vision models, the optimal design of prompt modules through a neural architecture search algorithm, specifically for each downstream dataset. By conducting extensive experiments on over 20 vision datasets, we demonstrate that NOAH (i) is superior to individual prompt modules, (ii) has a good few-shot learning ability, and (iii) is domain-generalizable. The code and models are available at https://github.com/Davidzhangyuanhan/NOAH.