Research map / Probabilistic and scientific modeling / Variational and generative sampling
Dataset Dynamics via Gradient Flows in Probability Space
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 |
|---|---|---|---|
| Flowing target dataset to MNIST with supervised OTDD gradient flow (labels used in flow) | frozen MNIST classifier on USPS, KMNIST, FashionMNIST, EMNIST · multi-dataset model re-purposing, 5K samples | accuracy (%): e.g. KMNIST 5.05 -> 99.2; FMNIST 8.22 -> 99.1; EMNIST only 3.62 -> 36.7 | Helped |
| Unsupervised flow with k-means pseudo-labels instead of true labels (ablation) | frozen MNIST classifier on USPS, KMNIST, FashionMNIST, EMNIST · multi-dataset model re-purposing, 5K samples | accuracy (%): improves on KMNIST/FMNIST/EMNIST, hurts on USPS (77.1 -> 66.4) | Mixed |
Closest papers by meaning
- Flowing Datasets with Wasserstein over Wasserstein Gradient Flows (ICML 2025 · similarity 0.92)
- Wasserstein Transfer Learning (NeurIPS 2025 · similarity 0.85)
- Diffusion Generative Flow Samplers: Improving learning signals through partial trajectory optimization (ICLR 2024 · similarity 0.84)
- Embarrassingly Simple Dataset Distillation (ICLR 2024 · similarity 0.84)
- Annealing Flow Generative Models Towards Sampling High-Dimensional and Multi-Modal Distributions (ICML 2025 · similarity 0.84)
- Generalized Wasserstein Flow Matching: Transport Plans, Everywhere, All at Once (NeurIPS 2026 · similarity 0.84)
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.