UniTTA: Unified Benchmark and Versatile Framework Towards Realistic Test-Time Adaptation

We propose a Unified Test-Time Adaptation ($\textbf{UniTTA}$) benchmark, which is comprehensive and widely applicable.Alongside this benchmark, we propose a versatile UniTTA framework, which includes a Balanced Domain Normalization (BDN) layer and a COrrelated Feature Adaptation (COFA) method--designed to mitigate distribution gaps in domain and class, respectively.Extensive experiments demonstrate that our framework excels within the UniTTA benchmark and achieves state-of-the-art performance on average.