Bidirectional Prototype-Reward co-Evolution for Test-Time Adaptation of Vision-Language Models

Re-assign community