Scaling and Distilling Text Embeddings for Better Diffusibility
Diffusion language models (DLMs) offer a promising alternative to autoregressive (AR) language generation. Recent advances in continuous DLMs, which apply latent diffusion to continuous text embeddings, raise a practical question: which embedding makes the best latent space, i.e., the most diffusible? To answer this, w...
Ze-Kai Zhang, Yun-Jie Tian, Yan-Jin He et al.
· 0 citations