AI RESEARCH

Guiding Distribution Matching Distillation with Gradient-Based Reinforcement Learning

arXiv CS.LG

ArXi:2604.19009v1 Announce Type: new Diffusion distillation, exemplified by Distribution Matching Distillation (DMD), has shown great promise in few-step generation but often sacrifices quality for sampling speed. While integrating Reinforcement Learning (RL) into distillation offers potential, a naive fusion of these two objectives relies on suboptimal raw sample evaluation. This sample-based scoring creates inherent conflicts with the distillation trajectory and produces unreliable rewards due to the noisy nature of early-stage generation.