AI RESEARCH
NEO: No-Optimization Test-Time Adaptation through Latent Re-Centering
arXiv CS.LG
•
ArXi:2510.05635v2 Announce Type: replace Test-Time Adaptation (TTA) methods are often computationally expensive, require a large amount of data for effective adaptation, or are brittle to hyperparameters. Based on a theoretical foundation of the geometry of the latent space, we are able to significantly improve the alignment between source and distribution-shifted samples by re-centering target data embeddings at the origin. This insight motivates NEO -- a hyperparameter-free fully TTA method, that adds no significant compute compared to vanilla inference.