Research map / Probabilistic and scientific modeling / Variational and generative sampling
Sliced-Wasserstein Flows: Nonparametric Generative Modeling via Optimal Transport and Diffusions
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 |
|---|---|---|---|
| Run SWF to transport particles toward the target distribution | SWF · 2D Gaussian mixture model | sliced Wasserstein cost | Helped |
| Reuse training-stage projected quantile functions for new particles | SWF · 2D Gaussian mixture model test stage | sliced Wasserstein cost | Hurt |
| Increase the entropy regularization parameter λ (ablation) | SWF · 2D Gaussian mixture model | particle spread | Helped |
| Transport samples in autoencoder bottleneck space | SWF · MNIST bottleneck features | sample quality | Helped |
| Transport samples in autoencoder bottleneck space | SWF · CelebA bottleneck features | sample quality | Helped |
| Reduce dimension using an autoencoder instead of operating in the original image space (ablation) | SWF · MNIST | convergence speed | Helped |
| Reduce dimension using an autoencoder instead of operating in the original image space (ablation) | SWF · CelebA | convergence speed | Helped |
| Keep projection directions fixed throughout iterations (ablation) | SWF · MNIST original image space | sample quality | Hurt |
| Resample projection directions at each iteration (ablation) | SWF · high-dimensional image data | convergence speed | Helped |
Closest papers by meaning
- Scalable Wasserstein Gradient Flow for Generative Modeling through Unbalanced Optimal Transport (ICML 2024 · similarity 0.90)
- Flowing Datasets with Wasserstein over Wasserstein Gradient Flows (ICML 2025 · similarity 0.89)
- Multimarginal Generative Modeling with Stochastic Interpolants (ICLR 2024 · similarity 0.88)
- Generalized Wasserstein Flow Matching: Transport Plans, Everywhere, All at Once (NeurIPS 2026 · similarity 0.88)
- Towards Non-Asymptotic Convergence for Diffusion-Based Generative Models (ICLR 2024 · similarity 0.88)
- Stochastic Interpolants with Data-Dependent Couplings (ICML 2024 · 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.