Diffusion models trained on large datasets can synthesize photo-realistic images of remarkable quality and diversity. However, attributing these images back to the training data-that is, identifying specific training examples which caused an image to be generated-remains a challenge. In this paper, we propose a framework that: (i) provides a formal notion of data attribution in the context of diffusion models, and (ii) allows us to counterfactually validate such attributions. Then, we provide a method for computing these attributions efficiently. Finally, we apply our method to find (and evaluate) such attributions for denoising diffusion probabilistic models trained on CIFAR-10 and latent diffusion models trained on MS COCO. We provide code at https://github.com/MadryLab/journey-TRAK .
@article{arxiv.2312.06205,
title = {The Journey, Not the Destination: How Data Guides Diffusion Models},
author = {Kristian Georgiev and Joshua Vendrow and Hadi Salman and Sung Min Park and Aleksander Madry},
journal= {arXiv preprint arXiv:2312.06205},
year = {2023}
}