Research map / Probabilistic and scientific modeling / Variational and generative sampling
Generative Modeling by Estimating Gradients of the Data Distribution
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use a single noise level (σ=0.01) with vanilla Langevin dynamics instead of multiple noise levels and annealed Langevin dynamics (ablation) | NCSN image generation · MNIST, CelebA, and CIFAR-10 | sample quality | Hurt |
| Add small Gaussian noise to CIFAR-10 data before sliced score matching (ablation) | ResNet score estimation · CIFAR-10 | sliced score matching loss convergence | Helped |
| Use sliced score matching rather than denoising score matching to train NCSNs (ablation) | NCSN · noise-perturbed data distributions | score estimation performance | No clear effect |
| Use annealed Langevin dynamics rather than standard Langevin dynamics (ablation) | score-based sampling · two-mode Gaussian mixture | recovery of relative mode weights | Helped |
| Use NCSN for unconditional CIFAR-10 generation instead of ProgressiveGAN | CIFAR-10 unconditional generation · CIFAR-10 unconditional | Inception Score: 8.80±.05 -> 8.87 ± .12 | Helped |
| Use NCSN for unconditional CIFAR-10 generation instead of MoLM | CIFAR-10 unconditional generation · CIFAR-10 unconditional | FID: 18.9 -> 25.32 | Hurt |
Closest papers by meaning
- Adversarial score matching and improved sampling for image generation (ICLR 2021 · similarity 0.90)
- Statistically Optimal Generative Modeling with Maximum Deviation from the Empirical Distribution (ICML 2024 · similarity 0.89)
- Variational Annealing of GANs: A Langevin Perspective (ICML 2019 · similarity 0.89)
- Score-Based Generative Modeling with Critically-Damped Langevin Diffusion (ICLR 2022 · similarity 0.89)
- Adaptive Density Estimation for Generative Models (NeurIPS 2019 · similarity 0.89)
- Revisiting Diffusion Models: From Generative Pre-training to One-Step Generation (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.