Empowering Diffusion Models on the Embedding Space for Text Generation

arXiv:2212.09412v3 Announce Type: replace-cross
Abstract: Diffusion models have achieved state-of-the-art synthesis quality on both visual and audio tasks, and recent works further adapt them to textual data by diffusing on the embedding space. In this paper, we conduct systematic studies of the optimization challenges encountered with both the embedding space and the denoising model, which have not been carefully explored. Firstly, the data distribution is learnable for embeddings, which may lead to the collapse of the embedding space and unstable …

abstract adapt art arxiv audio challenges cs.ai cs.cl cs.lg data diffusion diffusion models embedding optimization paper quality space state studies synthesis tasks text text generation textual them type visual