AI RESEARCH
KLASS: KL-Guided Fast Inference in Masked Diffusion Models
arXiv CS.LG
•
ArXi:2511.05664v2 Announce Type: replace Masked diffusion models have nstrated competitive results on various tasks including language generation. However, due to its iterative refinement process, the inference is often bottlenecked by slow and static sampling speed. To overcome this problem, we