# Uncertainty-Aware Prototype Learning for Open-Set Test-Time Adaptation

DevFeed: [Uncertainty-Aware Prototype Learning for Open-Set Test-Time Adaptation](<https://devfeed.tech/articles/uncertainty-aware-prototype-learning-for-open-set-test-time-adaptation-116972.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.04689>)

Author: Jinwu Hu, Langyu Bian, Jian Sun, Dongjin Yang, Chao Shuai, Yufeng Wang, Kai Zhou, Bin Xiao, Mingkui Tan, Hongmin Cai, Wei Luo

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>), [outlier-detection](<https://devfeed.tech/topics/outlier-detection.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [c](<https://devfeed.tech/tags/c.md>), [class](<https://devfeed.tech/tags/class.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [entropy](<https://devfeed.tech/tags/entropy.md>), [features](<https://devfeed.tech/tags/features.md>), [learning](<https://devfeed.tech/tags/learning.md>), [ood](<https://devfeed.tech/tags/ood.md>), [open](<https://devfeed.tech/tags/open.md>), [performance](<https://devfeed.tech/tags/performance.md>), [prototype](<https://devfeed.tech/tags/prototype.md>), [test](<https://devfeed.tech/tags/test.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>)

## AI overview

UAPA addresses open-set test-time adaptation, where shifted in-distribution (csID) and out-of-distribution (csOOD) samples can appear together and csOOD samples may cause harmful adaptation gradients. It separates them using a classifier-prototype score and a mini-batch-normalized, evidence-derived uncertainty score. It periodically refines class-specific target prototypes from buffered csID features, then minimizes entropy for csID and maximizes it for csOOD; prototype-level contrastive learning pulls csID features toward matching prototypes and limits csOOD similarity to known-class prototypes. The abstract reports average OSCR gains over the strongest baselines of 2.08 percentage points on CIFAR benchmarks and 1.28 points on Tiny-ImageNet-C.

## Source excerpt

arXiv:2611.04689v1 Announce Type: new Abstract: Open-set test-time adaptation (TTA) aims to adapt source-trained models to unlabeled test streams where covariate-shifted in-distribution (csID) samples and covariate-shifted out-of-distribution (csOOD) samples may coexist. Existing TTA methods are typically designed under a closed-set assumption and therefore lack explicit mechanisms for reliably distinguishing csID from csOOD samples. Consequently, csOOD samples may be mistakenly included in confidence-enhancing adaptation, producing harmful gradients that degrade known-class performance. To address this, we propose Uncertainty-Aware Prototype Learning for Open-Set Test-Time Adaptation (UAPA), a plug-and-play method for reliable open-set adaptation. Specifically, we first introduce Uncertainty-Driven ID/OOD Separation, which combines a classifier-prototype score with a mini-batch-normalized evidence-derived uncertainty score to distinguish csID from csOOD samples. We then propose a Dynamic Prototype Refinement strategy that periodically refines class-specific target prototypes using buffered csID features to progressively follow the estimated class directions. Lastly, we introduce Contrastive Prototype Entropy Adaptation, which applies entropy minimization to csID samples and entropy maximization to csOOD samples, while prototype-level contrastive learning aligns csID features with their corresponding target prototypes and reduces the maximum similarity between csOOD features and known-class prototypes. Empirically, UAPA outperforms the strongest baselines on the CIFAR benchmarks and Tiny-ImageNet-C, improving average OSCR by 2.08 and 1.28 percentage points, respectively.