AI RESEARCH

FOZO: Forward-Only Zeroth-Order Prompt Optimization for Test-Time Adaptation

arXiv CS.CV

ArXi:2603.04733v2 Announce Type: replace Test-Time Adaptation (TTA) is essential for enabling deep learning models to handle real-world data distribution shifts. However, current approaches face significant limitations: backpropagation-based methods are not suitable for low-end deployment devices, due to their high computation and memory requirements, as well as their tendency to modify model weights during adaptation; while traditional backpropagation-free techniques exhibit constrained adaptation capabilities.