AcTTA: Rethinking Test-Time Adaptation via Dynamic Activation

TL;DR

AcTTA introduces a novel activation function reparameterization approach for test-time adaptation, enabling models to adapt to domain shifts without altering weights or using source data.

Contribution

It reinterprets activation functions as learnable parameters, allowing adaptive adjustment during inference to improve robustness under distribution shifts.

Findings

AcTTA outperforms normalization-based TTA methods on CIFAR-C, CIFAR100-C, and ImageNet-C.

Activation adaptation provides a compact, effective alternative for domain-shift robustness.

The method achieves stable and robust performance without modifying network weights.

Abstract

Test-time adaptation (TTA) aims to mitigate performance degradation under distribution shifts by updating model parameters during inference. Existing approaches have primarily framed adaptation around affine modulation, focusing on recalibrating normalization layers. This perspective, while effective, overlooks another influential component in representation dynamics: the activation function. We revisit this overlooked space and propose AcTTA, an activation-aware framework that reinterprets conventional activation functions from a learnable perspective and updates them adaptively at test time. AcTTA reformulates conventional activation functions (e.g., ReLU, GELU) into parameterized forms that shift their response threshold and modulate gradient sensitivity, enabling the network to adjust activation behavior under domain shifts. This functional reparameterization enables continuous adjustment of activation behavior without modifying network weights or requiring source data. Despite its simplicity, AcTTA achieves robust and stable adaptation across diverse corruptions. Across CIFAR10-C, CIFAR100-C, and ImageNet-C, AcTTA consistently surpasses normalization-based TTA methods. Our findings highlight activation adaptation as a compact and effective route toward domain-shift-robust test-time learning, broadening the prevailing affine-centric view of adaptation.