Neural Network Reprogrammability: A Unified Theme on Model Reprogramming, Prompt Tuning, and Prompt Instruction

TL;DR

This paper introduces neural network reprogrammability as a unifying framework for various model adaptation techniques, highlighting their shared principles and categorizing them across key dimensions to advance efficient model customization.

Contribution

It provides a comprehensive taxonomy and theoretical unification of model reprogramming, prompt tuning, and prompt instruction, connecting fragmented research areas under a common paradigm.

Findings

Reprogramming techniques manipulate model interfaces without changing parameters.

A taxonomy categorizes adaptation methods across four key dimensions.

Framework applies across data modalities and model architectures.

Abstract

As large-scale pre-trained foundation models continue to expand in size and capability, efficiently adapting them to specific downstream tasks has become increasingly critical. Despite substantial progress, existing adaptation approaches have evolved largely in isolation, without a clear understanding of their interrelationships. This survey introduces neural network reprogrammability as a unifying framework that bridges mainstream model adaptation techniques--model reprogramming, prompt tuning, and prompt instruction--previously fragmented research areas yet converges on a shared principle: repurposing a pre-trained model by manipulating information at the interfaces while keeping the model parameters frozen. These methods exploit neural networks' sensitivity to manipulation on different interfaces, be it through perturbing inputs, inserting tokens into intermediate layers, or providing task-specific examples in context, to redirect model behaviors towards desired outcomes. We then present a taxonomy that categorizes such information manipulation-based adaptation approaches across four key dimensions: manipulation format (fixed or learnable), location (interfaces where manipulations occur), operator (how they are applied), and output alignment requirement (post-processing needed to align outputs with downstream tasks). Notably, this framework applies consistently across data modalities, independent of specific model architectures. Moreover, viewing established techniques like in-context learning and chain-of-thought prompting through this lens reveals both their theoretical connections and practical distinctions. We further analyze remaining technical challenges and ethical considerations, positioning neural network reprogrammability as a fundamental paradigm for efficient model adaptation. We lastly identify promising research directions emerging from this integrative viewpoint.