It looks like they use their own dataset and should be able to avoid that kind of information leakage. From their paper: "Our base diffusion model is a re-imp

It looks like they use their own dataset and should be able to avoid that kind of information leakage. From their paper: "Our base diffusion model is a re-implementation of the Latent Diffusion Model [42] that has been trained on an internal dataset of image-text pairs with input resolution 512×512×3 " They also use multi-view datasets during training, but presumably they haven't included those in the diffusion pretraining. "Training Dataset To learn a generalizable diffusion prior for novel view synthesis, we train on a mixture of the synthetic Objaverse [10] dataset and three real-world datasets: CO3D [38], MVImgNet [64], and RealEstate10K"