Research map / Image generation / Fast diffusion sampling
Data-to-Energy Stochastic Dynamics
Kirill Tamogashev, Nikolay Malkin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| learn diffusion variance in addition to drift (ablation) | data-to-data IPF (log-likelihood) · Two Moons<->GMM (K=20) | W2^2: 0.031 -> 0.023 | Improved |
| learn diffusion variance in addition to drift (ablation) | data-to-data IPF (log-likelihood) · Gauss<->Two Moons (K=20) | W2^2: 0.033 -> 0.022 | Improved |
| learn diffusion variance (ablation) | data-to-data IPF · Gauss<->GMM (K=20) | W2^2: 0.037 -> 0.042 | Hurt |
| learn diffusion variance (ablation) | data-to-data IPF · Gauss<->GMM (K=5) | W2^2: 0.066 -> 0.026 | Improved |
| learn diffusion variance (ablation) | data-to-data IPF · Gauss<->GMM (K=5) | Path KL: 0.988 -> 4.158 | Hurt |
| learn diffusion variance with large step count (ablation) | data-to-data IPF · Gauss<->GMM (K=40) | W2^2: 0.023 -> 0.028 | No clear effect |
| add replay buffer to on-policy training (ablation) | data-to-energy IPF · CIFAR-10 SN-GAN dogs | Path KL: 1506.407 -> 622.895 | Improved |
| add Langevin updates on buffer (ablation) | data-to-energy IPF with buffer · CIFAR-10 SN-GAN dogs | Path KL: 622.895 -> 383.514 | Improved |
| reuse backward trajectories in loss (ablation) | data-to-energy IPF with buffer and Langevin · CIFAR-10 SN-GAN dogs | Path KL: 383.514 -> 206.094 | Improved |
| reuse backward trajectories in loss (ablation) | data-to-energy IPF with buffer and Langevin · CIFAR-10 SN-GAN dogs | mean log-reward: -0.286 -> -0.657 | Hurt |
| use smaller off-policy ratio (ablation) | data-to-energy IPF · CIFAR-10 SN-GAN dogs | Path KL: 206.094 -> 668.255 | Hurt |
| anneal off-policy ratio (ablation) | data-to-energy IPF · CIFAR-10 SN-GAN dogs | mean log-reward: -0.657 -> -0.149 | Mixed |
Closest papers by meaning
- Learning non-equilibrium diffusions with Schrödinger bridges: from exactly solvable to simulation-free (NeurIPS 2025 · similarity 0.93)
- Go With the Flow: Fast Diffusion for Gaussian Mixture Models (NeurIPS 2025 · similarity 0.92)
- Variational Schrödinger Diffusion Models (ICML 2024 · similarity 0.92)
- Generalized Schrödinger Bridge Matching (ICLR 2024 · similarity 0.92)
- Light and Optimal Schrödinger Bridge Matching (ICML 2024 · similarity 0.91)
- Momentum Multi-Marginal Schrödinger Bridge Matching (NeurIPS 2025 · similarity 0.90)
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.
