Active Prompt Learning with Vision-Language Model Priors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Hoyoung, Jin, Seokhee, Sung, Changhwan, Kim, Jaechang, Ok, Jungseul
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908874575970304
author Kim, Hoyoung
Jin, Seokhee
Sung, Changhwan
Kim, Jaechang
Ok, Jungseul
author_facet Kim, Hoyoung
Jin, Seokhee
Sung, Changhwan
Kim, Jaechang
Ok, Jungseul
contents Vision-language models (VLMs) have demonstrated remarkable zero-shot performance across various classification tasks. Nonetheless, their reliance on hand-crafted text prompts for each task hinders efficient adaptation to new tasks. While prompt learning offers a promising solution, most studies focus on maximizing the utilization of given few-shot labeled datasets, often overlooking the potential of careful data selection strategies, which enable higher accuracy with fewer labeled data. This motivates us to study a budget-efficient active prompt learning framework. Specifically, we introduce a class-guided clustering that leverages the pre-trained image and text encoders of VLMs, thereby enabling our cluster-balanced acquisition function from the initial round of active learning. Furthermore, considering the substantial class-wise variance in confidence exhibited by VLMs, we propose a budget-saving selective querying based on adaptive class-wise thresholds. Extensive experiments in active learning scenarios across seven datasets demonstrate that our method outperforms existing baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2411_16722
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Active Prompt Learning with Vision-Language Model Priors
Kim, Hoyoung
Jin, Seokhee
Sung, Changhwan
Kim, Jaechang
Ok, Jungseul
Computer Vision and Pattern Recognition
Vision-language models (VLMs) have demonstrated remarkable zero-shot performance across various classification tasks. Nonetheless, their reliance on hand-crafted text prompts for each task hinders efficient adaptation to new tasks. While prompt learning offers a promising solution, most studies focus on maximizing the utilization of given few-shot labeled datasets, often overlooking the potential of careful data selection strategies, which enable higher accuracy with fewer labeled data. This motivates us to study a budget-efficient active prompt learning framework. Specifically, we introduce a class-guided clustering that leverages the pre-trained image and text encoders of VLMs, thereby enabling our cluster-balanced acquisition function from the initial round of active learning. Furthermore, considering the substantial class-wise variance in confidence exhibited by VLMs, we propose a budget-saving selective querying based on adaptive class-wise thresholds. Extensive experiments in active learning scenarios across seven datasets demonstrate that our method outperforms existing baselines.
title Active Prompt Learning with Vision-Language Model Priors
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.16722