Test-Time Adaptation of Vision-Language Models for Open-Vocabulary Semantic Segmentation - View it on GitHub