OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hua, Cong, Xu, Qianqian, Yang, Zhiyong, Wang, Zitai, Bao, Shilong, Huang, Qingming
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910932840480768
author Hua, Cong
Xu, Qianqian
Yang, Zhiyong
Wang, Zitai
Bao, Shilong
Huang, Qingming
author_facet Hua, Cong
Xu, Qianqian
Yang, Zhiyong
Wang, Zitai
Bao, Shilong
Huang, Qingming
contents Prompt tuning adapts Vision-Language Models like CLIP to open-world tasks with minimal training costs. In this direction, one typical paradigm evaluates model performance separately on known classes (i.e., base domain) and unseen classes (i.e., new domain). However, real-world scenarios require models to handle inputs without prior domain knowledge. This practical challenge has spurred the development of open-world prompt tuning, which demands a unified evaluation of two stages: 1) detecting whether an input belongs to the base or new domain (P1), and 2) classifying the sample into its correct class (P2). What's more, as domain distributions are generally unknown, a proper metric should be insensitive to varying base/new sample ratios (P3). However, we find that current metrics, including HM, overall accuracy, and AUROC, fail to satisfy these three properties simultaneously. To bridge this gap, we propose OpenworldAUC, a unified metric that jointly assesses detection and classification through pairwise instance comparisons. To optimize OpenworldAUC effectively, we introduce Gated Mixture-of-Prompts (GMoP), which employs domain-specific prompts and a gating mechanism to dynamically balance detection and classification. Theoretical guarantees ensure generalization of GMoP under practical conditions. Experiments on 15 benchmarks in open-world scenarios show GMoP achieves SOTA performance on OpenworldAUC and other metrics. We release the code at https://github.com/huacong/OpenworldAUC
format Preprint
id arxiv_https___arxiv_org_abs_2505_05180
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning
Hua, Cong
Xu, Qianqian
Yang, Zhiyong
Wang, Zitai
Bao, Shilong
Huang, Qingming
Machine Learning
Prompt tuning adapts Vision-Language Models like CLIP to open-world tasks with minimal training costs. In this direction, one typical paradigm evaluates model performance separately on known classes (i.e., base domain) and unseen classes (i.e., new domain). However, real-world scenarios require models to handle inputs without prior domain knowledge. This practical challenge has spurred the development of open-world prompt tuning, which demands a unified evaluation of two stages: 1) detecting whether an input belongs to the base or new domain (P1), and 2) classifying the sample into its correct class (P2). What's more, as domain distributions are generally unknown, a proper metric should be insensitive to varying base/new sample ratios (P3). However, we find that current metrics, including HM, overall accuracy, and AUROC, fail to satisfy these three properties simultaneously. To bridge this gap, we propose OpenworldAUC, a unified metric that jointly assesses detection and classification through pairwise instance comparisons. To optimize OpenworldAUC effectively, we introduce Gated Mixture-of-Prompts (GMoP), which employs domain-specific prompts and a gating mechanism to dynamically balance detection and classification. Theoretical guarantees ensure generalization of GMoP under practical conditions. Experiments on 15 benchmarks in open-world scenarios show GMoP achieves SOTA performance on OpenworldAUC and other metrics. We release the code at https://github.com/huacong/OpenworldAUC
title OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning
topic Machine Learning
url https://arxiv.org/abs/2505.05180