Paper recorded by Signals 4 on 2026-09-30 in cs.LG. Abstract reproduced from arXiv; link to the original below.
Published 2026-09-30 on arXiv · recorded by Signals 4 on 2026-10-01
Category: cs.LG · 机器学习 · first seen 2026-10-01
Continuous diffusion language models generate all tokens in parallel, yet high-quality generation can still require hundreds of network evaluations (NFEs). We study how distributional distillation can reduce this cost by exploiting the student's probabilistic token outputs. Our unified formulation connects the student's output parameterization to the resulting gradient estimators and yields two me