:nerd: Our new preprint on interpretable evaluation for text-to-image and text-to-video generative models is now available! Link to Arxiv