Intriguing Properties of Data Attribution on Diffusion Models

arXiv:2311.00500v2 Announce Type: replace
Abstract: Data attribution seeks to trace model outputs back to training data. With the recent development of diffusion models, data attribution has become a desired module to properly assign valuations for high-quality or copyrighted training samples, ensuring that data contributors are fairly compensated or credited. Several theoretically motivated methods have been proposed to implement data attribution, in an effort to improve the trade-off between computational scalability and effectiveness. In this work, we conduct extensive experiments and …

arxiv attribution cs.ai cs.cv cs.lg data diffusion diffusion models type