Skip to content
Conference Open access

Efficient Test-Time Adaptation for Dynamic Distribution Shifts

Aug 2026 · 2026 6th International Conference on Advanced Algorithms and Neural Networks (AANN) · pp. 1-7 · 0 citations · 15 references

TL;DR

This paper presents an efficient online TTA framework for dynamic distribution shifts that couples reliability-dependent sample admission, residual branch-only updating, and statistics-level temporal memory within the same fixed online update budget.

Abstract

Pre-trained visual models are increasingly deployed in streaming environments where the target distribution changes with weather, illumination, sensor aging, background composition, and time. Existing test-time adaptation (TTA) algorithms usually assume either a fixed target domain or sufficiently large independent test batches; consequently, they can be computationally heavy, slow to converge, and vulnerable to noisy pseudo-supervision under dynamic shifts. This paper presents an efficient online TTA framework for dynamic distribution shifts. The framework contains dynamic confidence screening, branch-level incremental tuning, and temporal distribution smoothness. Only a small set of adaptive branches is updated during inference, while unreliable samples are filtered before optimization and historical batch statistics regularize short-term drift. Experiments on dynamic image classification and object detection streams show average accuracy gains of 4.2%–8.7% over representative TTA baselines and more than 60% lower inference latency than fullparameter continual adaptation. The proposed design provides a practical balance between robustness, stability, and real-time deployment efficiency. Unlike prior TTA designs that treat filtering, lightweight updating, or temporal stabilization separately, the proposed framework couples reliability-dependent sample admission, residual branch-only updating, and statistics-level temporal memory within the same fixed online update budget. Because the classification corruption streams are manually scheduled, the evidence is interpreted as controlled dynamic-benchmark validation rather than a substitute for field deployment.

Read PDF

Similar papers

Preprint Oct 2026

The Batch Size Can Be One: Test-Time Adaptation from Classifier Geometry

Test-time adaptation updates a deployed model on an unlabeled test stream. In practice this stream often arrives one sample at a time, with the classes in no particular order. A single sample then provides limited information for estimating distribution-level statistics, and selecting samples by confidence can admit pr...

J. Hwang, Yujin Oh, Juhyeon Shin et al. · 0 citations

From Scores to Samples: Elastic Forcing for Autoregressive Video Generation

This framework minimizes maximum mean discrepancy in frozen self-supervised video representation spaces in frozen self-supervised video representation spaces, using a hybrid Nystr\"om--Monte Carlo estimator to balance approximation bias and sampling variance.

Chi Zhang, Yue-Yi Liu, Hao-Yan Shi et al. · 3 citations · ⚡2
Preprint Oct 2026

Inference-Time Prior Adaptation for Imbalanced Data Streams under Concept Drift

Data stream classification under concept drift and class imbalance requires models to adapt continuously while maintaining strong minority-class performance. Many existing approaches rely on active memory management, in- cluding instance removal, memory restructuring, resampling, and synthetic data generation. We prese...

Tobias Callies, Eirini Ntoutsi, À. Munich et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.