Tech is getting smarter at spotting things in new places. Imagine you teach a computer to recognize cars using pictures from one city. Now, you want it to work in a different city with different lighting and roads. This is tricky because the computer might not recognize cars as well in the new place. One way to help the computer is to have it label some pictures its...

The goal of domain-adaptive object detection (DAOD) is to help detectors work well in new, unlabeled areas by reducing the bias that comes from the original, labeled areas. Recent work has shown that vision-language models (VLMs) can be very helpful in this process. However, there are some issues that need to be addressed. Most existing methods focus on single-domain...