Ranked Entropy Minimization for Continual Test-Time Adaptation

Jisu Han, Jaemin Na, Wonjun Hwang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:21798-21813, 2025.

Abstract

Test-time adaptation aims to adapt to realistic environments in an online manner by learning during test time. Entropy minimization has emerged as a principal strategy for test-time adaptation due to its efficiency and adaptability. Nevertheless, it remains underexplored in continual test-time adaptation, where stability is more important. We observe that the entropy minimization method often suffers from model collapse, where the model converges to predicting a single class for all images due to a trivial solution. We propose ranked entropy minimization to mitigate the stability problem of the entropy minimization method and extend its applicability to continuous scenarios. Our approach explicitly structures the prediction difficulty through a progressive masking strategy. Specifically, it gradually aligns the model’s probability distributions across different levels of prediction difficulty while preserving the rank order of entropy. The proposed method is extensively evaluated across various benchmarks, demonstrating its effectiveness through empirical results.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-han25e, title = {Ranked Entropy Minimization for Continual Test-Time Adaptation}, author = {Han, Jisu and Na, Jaemin and Hwang, Wonjun}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {21798--21813}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/han25e/han25e.pdf}, url = {https://proceedings.mlr.press/v267/han25e.html}, abstract = {Test-time adaptation aims to adapt to realistic environments in an online manner by learning during test time. Entropy minimization has emerged as a principal strategy for test-time adaptation due to its efficiency and adaptability. Nevertheless, it remains underexplored in continual test-time adaptation, where stability is more important. We observe that the entropy minimization method often suffers from model collapse, where the model converges to predicting a single class for all images due to a trivial solution. We propose ranked entropy minimization to mitigate the stability problem of the entropy minimization method and extend its applicability to continuous scenarios. Our approach explicitly structures the prediction difficulty through a progressive masking strategy. Specifically, it gradually aligns the model’s probability distributions across different levels of prediction difficulty while preserving the rank order of entropy. The proposed method is extensively evaluated across various benchmarks, demonstrating its effectiveness through empirical results.} }
Endnote
%0 Conference Paper %T Ranked Entropy Minimization for Continual Test-Time Adaptation %A Jisu Han %A Jaemin Na %A Wonjun Hwang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-han25e %I PMLR %P 21798--21813 %U https://proceedings.mlr.press/v267/han25e.html %V 267 %X Test-time adaptation aims to adapt to realistic environments in an online manner by learning during test time. Entropy minimization has emerged as a principal strategy for test-time adaptation due to its efficiency and adaptability. Nevertheless, it remains underexplored in continual test-time adaptation, where stability is more important. We observe that the entropy minimization method often suffers from model collapse, where the model converges to predicting a single class for all images due to a trivial solution. We propose ranked entropy minimization to mitigate the stability problem of the entropy minimization method and extend its applicability to continuous scenarios. Our approach explicitly structures the prediction difficulty through a progressive masking strategy. Specifically, it gradually aligns the model’s probability distributions across different levels of prediction difficulty while preserving the rank order of entropy. The proposed method is extensively evaluated across various benchmarks, demonstrating its effectiveness through empirical results.
APA
Han, J., Na, J. & Hwang, W.. (2025). Ranked Entropy Minimization for Continual Test-Time Adaptation. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:21798-21813 Available from https://proceedings.mlr.press/v267/han25e.html.

Related Material