Visual prompting reimagined: The power of the Activation Prompts

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Yihua, Li, Hongkang, Yao, Yuguang, Chen, Aochuan, Zhang, Shuai, Chen, Pin-Yu, Wang, Meng, Liu, Sijia
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915921888542720
author Zhang, Yihua
Li, Hongkang
Yao, Yuguang
Chen, Aochuan
Zhang, Shuai
Chen, Pin-Yu
Wang, Meng
Liu, Sijia
author_facet Zhang, Yihua
Li, Hongkang
Yao, Yuguang
Chen, Aochuan
Zhang, Shuai
Chen, Pin-Yu
Wang, Meng
Liu, Sijia
contents Visual prompting (VP) has emerged as a popular method to repurpose pretrained vision models for adaptation to downstream tasks. Unlike conventional model fine-tuning techniques, VP introduces a universal perturbation directly into the input data to facilitate task-specific fine-tuning rather than modifying model parameters. However, there exists a noticeable performance gap between VP and conventional fine-tuning methods, highlighting an unexplored realm in theory and practice to understand and advance the input-level VP to reduce its current performance gap. Towards this end, we introduce a generalized concept, termed activation prompt (AP), which extends the scope of the input-level VP by enabling universal perturbations to be applied to activation maps within the intermediate layers of the model. By using AP to revisit the problem of VP and employing it as an analytical tool, we demonstrate the intrinsic limitations of VP in both performance and efficiency, revealing why input-level prompting may lack effectiveness compared to AP, which exhibits a model-dependent layer preference. We show that AP is closely related to normalization tuning in convolutional neural networks and vision transformers, although each model type has distinct layer preferences for prompting. We also theoretically elucidate the rationale behind such a preference by analyzing global features across layers. Through extensive experiments across 29 datasets and various model architectures, we provide a comprehensive performance analysis of AP, comparing it with VP and parameter-efficient fine-tuning baselines. Our results demonstrate AP's superiority in both accuracy and efficiency, considering factors such as time, parameters, memory usage, and throughput.
format Preprint
id arxiv_https___arxiv_org_abs_2604_06440
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Visual prompting reimagined: The power of the Activation Prompts
Zhang, Yihua
Li, Hongkang
Yao, Yuguang
Chen, Aochuan
Zhang, Shuai
Chen, Pin-Yu
Wang, Meng
Liu, Sijia
Computer Vision and Pattern Recognition
Machine Learning
Visual prompting (VP) has emerged as a popular method to repurpose pretrained vision models for adaptation to downstream tasks. Unlike conventional model fine-tuning techniques, VP introduces a universal perturbation directly into the input data to facilitate task-specific fine-tuning rather than modifying model parameters. However, there exists a noticeable performance gap between VP and conventional fine-tuning methods, highlighting an unexplored realm in theory and practice to understand and advance the input-level VP to reduce its current performance gap. Towards this end, we introduce a generalized concept, termed activation prompt (AP), which extends the scope of the input-level VP by enabling universal perturbations to be applied to activation maps within the intermediate layers of the model. By using AP to revisit the problem of VP and employing it as an analytical tool, we demonstrate the intrinsic limitations of VP in both performance and efficiency, revealing why input-level prompting may lack effectiveness compared to AP, which exhibits a model-dependent layer preference. We show that AP is closely related to normalization tuning in convolutional neural networks and vision transformers, although each model type has distinct layer preferences for prompting. We also theoretically elucidate the rationale behind such a preference by analyzing global features across layers. Through extensive experiments across 29 datasets and various model architectures, we provide a comprehensive performance analysis of AP, comparing it with VP and parameter-efficient fine-tuning baselines. Our results demonstrate AP's superiority in both accuracy and efficiency, considering factors such as time, parameters, memory usage, and throughput.
title Visual prompting reimagined: The power of the Activation Prompts
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2604.06440