Geonhui Han

2 accepted papers

2026

AcTTA: Rethinking Test-Time Adaptation via Dynamic Activation

CVPR 2026

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 effectiv

Cited by 0 Source ↗Scholar ↗

2025

Buffer layers for Test-Time Adaptation

Hyeongyu Kim, GeonHui Han, Dosik Hwang

NeurIPS 2025poster

In recent advancements in Test Time Adaptation (TTA), most existing methodologies focus on updating normalization layers to adapt to the test domain. However, the reliance on normalization-based adaptation presents key challenges. First, normalization layers such as Batch Normalization (BN) are high…

Cited by 0 Source ↗code ↗Scholar ↗