WISE-ATTA: When to Ask for Labels in Budgeted Active Test-Time Adaptation

Hugging Face Daily Papers Papers
active-test-time-adaptation domain-shift label-efficiency distribution-shift test-time-adaptation model-robustness

Summary

WISE-ATTA introduces budgeted active test-time adaptation, deciding when to query labels over a test stream using lightweight online signals and a drift-based sample selection criterion, achieving competitive results with substantially fewer labels on ImageNet-C and ImageNet-R/K/A.

Active test-time adaptation (ATTA) improves robustness under distribution shift by updating a deployed model during inference while selectively querying supervision. However, most existing ATTA methods implicitly assume that supervision can be requested for every incoming test batch, which can incur substantial annotation cost over long test streams. In this work, we introduce budgeted ATTA in which labels are available for only a fraction of test batches. This formulation shifts the central challenge from deciding what to label within a batch to deciding when supervision should be applied over time. To address this challenge, we propose a budget-aware approach WISE-ATTA that allocates supervision over the test stream based on lightweight signals computed online, prioritizing periods where supervision is likely to be most useful. When a batch is selected for supervision, we further employ a drift-based sample selection criterion that targets samples exhibiting ongoing, unconverged adaptation dynamics, enabling effective updates from a single labeled example. We evaluate this approach on synthetic corruptions (ImageNet-C) and natural distribution shifts (ImageNet-R/K/A). Across settings, WISE-ATTA achieves competitive or improved performance compared to recent ATTA methods while requiring substantially fewer labels. Overall, we find that the timing of supervision is a key, yet underexplored, aspect of active test-time adaptation. Code: https://github.com/Muhammad-Huzaifaa/WISE-ATTA
Original Article
View Cached Full Text

Cached at: 09/30/26, 12:16 PM

Paper page - WISE-ATTA: When to Ask for Labels in Budgeted Active Test-Time Adaptation

Source: https://huggingface.co/papers/2609.37687

Abstract

Activetest-timeadaptation(ATTA)improvesrobustnessunderdistributionshiftbyupdatingadeployedmodelduringinferencewhileselectivelyqueryingsupervision.However,mostexistingATTAmethodsimplicitlyassumethatsupervisioncanberequestedforeveryincomingtestbatch,whichcanincursubstantialannotationcostoverlongteststreams.Inthiswork,weintroducebudgetedATTAinwhichlabelsareavailableforonlyafractionoftestbatches.Thisformulationshiftsthecentralchallengefromdecidingwhattolabelwithinabatchtodecidingwhensupervisionshouldbeappliedovertime.Toaddressthischallenge,weproposeabudget-awareapproachWISE-ATTAthatallocatessupervisionovertheteststreambasedonlightweightsignalscomputedonline,prioritizingperiodswheresupervisionislikelytobemostuseful.Whenabatchisselectedforsupervision,wefurtheremployadrift-basedsampleselectioncriterionthattargetssamplesexhibitingongoing,unconvergedadaptationdynamics,enablingeffectiveupdatesfromasinglelabeledexample.Weevaluatethisapproachonsyntheticcorruptions(ImageNet-C)andnaturaldistributionshifts(ImageNet-R/K/A).Acrosssettings,WISE-ATTAachievescompetitiveorimprovedperformancecomparedtorecentATTAmethodswhilerequiringsubstantiallyfewerlabels.Overall,wefindthatthetimingofsupervisionisakey,yetunderexplored,aspectofactivetest-timeadaptation.Code:https://github.com/Muhammad-Huzaifaa/WISE-ATTA

View arXiv pageView PDFGitHub3Add to collection

Get this paper in your agent:

hf papers read 2609\.37687

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.37687 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.37687 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.37687 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

AdaStop: Cost-Aware Early Stopping for DNN Test Selection

arXiv cs.LG

AdaStop is a cost-aware early stopping framework for DNN test selection that optimally stops labeling when the marginal fault discovery rate falls below a threshold, achieving 65-84% fault discovery using only 9-31% of the labeling budget.

REFIT: Recognize, Fix, and Test Wearable Sensor Placement Shifts without Labels

arXiv cs.LG

REFIT introduces a label-free input calibration method for frozen activity-recognition models that corrects wearable sensor placement shifts (e.g., switching wrists or reattaching straps) by fitting axis-transform families to match training statistics, outperforming existing test-time adaptation approaches without retraining.