Adaptive Debiasing Tsallis Entropy for Test-Time Adaptation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Xiangyu, Jiang, Dongming, Yu, Feng, Tian, Yueying, Tang, Jiaqi, Chen, Qing-Guo, Yang, Yang, Lu, Jianfeng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908829495590912
author Wu, Xiangyu
Jiang, Dongming
Yu, Feng
Tian, Yueying
Tang, Jiaqi
Chen, Qing-Guo
Yang, Yang
Lu, Jianfeng
author_facet Wu, Xiangyu
Jiang, Dongming
Yu, Feng
Tian, Yueying
Tang, Jiaqi
Chen, Qing-Guo
Yang, Yang
Lu, Jianfeng
contents Mainstream Test-Time Adaptation (TTA) methods for adapting vision-language models, e.g., CLIP, typically rely on Shannon Entropy (SE) at test time to measure prediction uncertainty and inconsistency. However, since CLIP has a built-in bias from pretraining on highly imbalanced web-crawled data, SE inevitably results in producing biased estimates of uncertainty entropy. To address this issue, we notably find and demonstrate that Tsallis Entropy (TE), a generalized form of SE, is naturally suited for characterizing biased distributions by introducing a non-extensive parameter q, with the performance of SE serving as a lower bound for TE. Building upon this, we generalize TE into Adaptive Debiasing Tsallis Entropy (ADTE) for TTA, customizing a class-specific parameter q^l derived by normalizing the estimated label bias from continuously incoming test instances, for each category. This adaptive approach allows ADTE to accurately select high-confidence views and seamlessly integrate with a label adjustment strategy to enhance adaptation, without introducing distribution-specific hyperparameter tuning. Besides, our investigation reveals that both TE and ADTE can serve as direct, advanced alternatives to SE in TTA, without any other modifications. Experimental results show that ADTE outperforms state-of-the-art methods on ImageNet and its five variants, and achieves the highest average performance on 10 cross-domain benchmarks, regardless of the model architecture or text prompts used. Our code is available at https://github.com/Jinx630/ADTE.
format Preprint
id arxiv_https___arxiv_org_abs_2602_11743
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Adaptive Debiasing Tsallis Entropy for Test-Time Adaptation
Wu, Xiangyu
Jiang, Dongming
Yu, Feng
Tian, Yueying
Tang, Jiaqi
Chen, Qing-Guo
Yang, Yang
Lu, Jianfeng
Computer Vision and Pattern Recognition
Mainstream Test-Time Adaptation (TTA) methods for adapting vision-language models, e.g., CLIP, typically rely on Shannon Entropy (SE) at test time to measure prediction uncertainty and inconsistency. However, since CLIP has a built-in bias from pretraining on highly imbalanced web-crawled data, SE inevitably results in producing biased estimates of uncertainty entropy. To address this issue, we notably find and demonstrate that Tsallis Entropy (TE), a generalized form of SE, is naturally suited for characterizing biased distributions by introducing a non-extensive parameter q, with the performance of SE serving as a lower bound for TE. Building upon this, we generalize TE into Adaptive Debiasing Tsallis Entropy (ADTE) for TTA, customizing a class-specific parameter q^l derived by normalizing the estimated label bias from continuously incoming test instances, for each category. This adaptive approach allows ADTE to accurately select high-confidence views and seamlessly integrate with a label adjustment strategy to enhance adaptation, without introducing distribution-specific hyperparameter tuning. Besides, our investigation reveals that both TE and ADTE can serve as direct, advanced alternatives to SE in TTA, without any other modifications. Experimental results show that ADTE outperforms state-of-the-art methods on ImageNet and its five variants, and achieves the highest average performance on 10 cross-domain benchmarks, regardless of the model architecture or text prompts used. Our code is available at https://github.com/Jinx630/ADTE.
title Adaptive Debiasing Tsallis Entropy for Test-Time Adaptation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.11743