Research map / Image generation / Fast diffusion sampling
Maximum Likelihood Training for Score-based Diffusion ODEs by High Order Denoising Score Matching
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 |
|---|---|---|---|
| Add second-order score matching (ablation) | VE · CIFAR-10 | NLL (bits/dim): 3.66 -> 3.44 | Helped |
| Add third-order score matching (ablation) | VE · CIFAR-10 | NLL (bits/dim): 3.66 -> 3.38 | Helped |
| Add second-order score matching (ablation) | VE · CIFAR-10 | FID: 2.42 -> 2.37 | Helped |
| Add third-order score matching (ablation) | VE · CIFAR-10 | FID: 2.42 -> 2.95 | Hurt |
| Add second-order score matching (ablation) | VE · ImageNet 32x32 | NLL (bits/dim): 4.21 -> 4.06 | Helped |
| Add third-order score matching (ablation) | VE · ImageNet 32x32 | NLL (bits/dim): 4.21 -> 4.04 | Helped |
| Add second-order score matching (ablation) | VE (deep) · CIFAR-10 | NLL (bits/dim): 3.45 -> 3.35 | Helped |
| Add third-order score matching (ablation) | VE (deep) · CIFAR-10 | NLL (bits/dim): 3.45 -> 3.27 | Helped |
| Add second-order score matching (ablation) | VE (deep) · CIFAR-10 | FID: 2.19 -> 2.43 | Hurt |
| Add third-order score matching (ablation) | VE (deep) · CIFAR-10 | FID: 2.19 -> 2.61 | Hurt |
| Add second-order score matching (ablation) | VE (deep) · ImageNet 32x32 | NLL (bits/dim): 4.21 -> 4.05 | Helped |
| Add third-order score matching (ablation) | VE (deep) · ImageNet 32x32 | NLL (bits/dim): 4.21 -> 4.03 | Helped |
Closest papers by meaning
- Training Diffusion-based Generative Models with Limited Data (ICML 2025 · similarity 0.91)
- What’s the score? Automated Denoising Score Matching for Nonlinear Diffusions (ICML 2024 · similarity 0.91)
- Neural Network-Based Score Estimation in Diffusion Models: Optimization and Generalization (ICLR 2024 · similarity 0.91)
- Nearest Neighbour Score Estimators for Diffusion Generative Models (ICML 2024 · similarity 0.90)
- Accelerating Convergence of Score-Based Diffusion Models, Provably (ICML 2024 · similarity 0.89)
- Dimension-free Score Matching and Time Bootstrapping for Diffusion Models (NeurIPS 2025 · similarity 0.89)
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.