Position: Do Not Explain Vision Models Without Context

Open in new window