[Collaborative Research]
LoRA can Replace Time and Class Embeddings in Diffusion Probabilistic Models
Joo Young Choi, Jaesung Park, Inkyu Park, Jaewoong Cho, Albert No, Ernest Ryu
Abstract
We propose LoRA modules as a replacement for the time and class embeddings of the U-Net architecture for diffusion probabilistic models. Our experiments on CIFAR-10 show that a score network trained with LoRA achieves competitive FID scores while being more efficient in memory compared to a score network trained with time and class embeddings.
- Vision & Animation
NeurIPS 2023 Workshop