Score Matching Model for Unbounded Data Score

By Dongjun Kim, Seungjae Shin, Kyungwoo Song, Wanmo Kang, Il-Chul Moon. 2021

32 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

Citations in the corpus

Score-based Generative Modeling in Latent Space (2021, background): “Finally, [72] proposed a modified time parameterization that allows modeling unbounded data scores.”
Score-Based Generative Modeling with Critically-Damped Langevin Diffusion (2021, methodology): “Currently used SDEs (Song et al., 2021c; Kim et al., 2021) have drift and diffusion coefficients of the simple form f(xt, t)=f(t)xt andG(xt, t)=g(t)Id. Generally, f and G are chosen such that the SDE’s marginal, equilibrium density is approximately Normal at time T , i.e., p(uT )≈N (0, Id).”
Fast Sampling of Diffusion Models with Exponential Integrator (2022, background): “This finding is not new and has been pointed out by existing works (Jolicoeur-Martineau et al., 2021; Kim et al., 2021; Song et al., 2020a).”
Tackling the Generative Learning Trilemma with Denoising Diffusion GANs (2021, unknown): “Table 4: Generative results on CelebA-HQ-256 Model FID↓ Denoising Diffusion GAN (ours) 7.64 Score SDE (Song et al., 2021c) 7.23 LSGM (Vahdat et al., 2021) 7.22 UDM (Kim et al., 2021) 7.16 NVAE (Vahdat & Kautz, 2020) 29.7 VAEBM (Xiao et al., 2021) 20.4 NCP-VAE (Aneja et al., 2021) 24.8 PGGAN (Karras ”
Score-based diffusion models for accelerated MRI (2021, unknown): “Score-based diffusion models are now one of the most popular methods in the context of image synthesis, matching the image fidelity of state-of-the-art GANs (Dhariwal and Nichol, 2021), and achieving the state-of-the-art log-likelihood on various datasets (Kingma et al., 2021; Kim et al., 2021).”