DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation
[Widzenie komputerowe] Autorzy: Zhengming Yu, Junkun Yuan, Haotian Yang i 8 więcej
Distribution Matching Distillation (DMD) trains a few-step student from the difference between separately estimated target and student scores, so it must keep an auxiliary diffusion model fitted to the student's evolving distribution at extra memory and computation cost. We introduce DMAD, Distribution Matching as Adversarial Distillation, which recasts distribution matching as classification and learns the required log-density ratios directly. Two discriminator heads on a shared backbone distinguish real data and teacher samples from the student's, and linear losses on their logits train the student without auxiliary score fitting. We prove that at the discriminator optimum these losses recover the distribution-matching gradient underlying DMD, through the classical identity linking discriminator logits to log-density ratios. We further introduce gap-based reweighting, which adapts tea…
- Kategoria arxiv
- cs.CV
- Kategoria
- Widzenie komputerowe