Explaining Computer Vision Models Requires Considering Spatial Context
Existing explainable AI (XAI) methods for computer vision models often fail to capture the importance of spatial context, which can be crucial for accurate predictions in real-world applications.