Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment

Young-Geun Choi · Sungeun Hong (Inha Univ.) · Youjia Zhang (Sungkyunkwan University) · Youngeun Kim (Amazon) · Hongyeob Kim (SungKyunKwan University) · Huiling Liu (Sung Kyun Kwan University)
backpropagation-freecalibrated predictionsclass-conditional feature distributionsclass-conditional likelihoodsclip priorsdecision boundariesdistribution shiftshistorical knowledge banklightweight regularizationlikelihood biasprobabilistic inferenceshared covariance matrixtest-time adaptationtraining-free inferencezero-shot robustness

Test-time adaptation (TTA) enhances the zero-shot robustness under distribution shifts by leveraging unlabeled test data during inference. Despite notable advances, several challenges still limit its broader applicability. First, most methods rely on backpropagation or iterative optimization, which limits scalability and hinders real-time deployment. Second, they lack explicit modeling of class-conditional feature distributions. This modeling is crucial for producing reliable decision boundaries and calibrated predictions, but it remains underexplored due to the lack of both source data and supervision at test time. In this paper, we propose ADAPT, an Advanced Distribution-Aware and backPropagation-free Test-time adaptation method. We reframe TTA as a probabilistic inference task by modeling class-conditional likelihoods using gradually updated class means and a shared covariance matrix. This enables closed-form, training-free inference. To correct potential likelihood bias, we introduce lightweight regularization guided by CLIP priors and a historical knowledge bank. ADAPT requires no source data, no gradient updates, and no full access to target data, supporting both online and transductive settings. Extensive experiments across diverse benchmarks demonstrate that our method achieves state-of-the-art performance under a wide range of distribution shifts with superior scalability and robustness.