Research map / Image generation / Fast diffusion sampling
User-defined Event Sampling and Uncertainty Quantification in Diffusion Models for Physical Dynamical Systems
Marc Anton Finzi, Anudhyan Boral, Andrew Gordon Wilson, Fei Sha, Leonardo Zepeda-Nunez
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 |
|---|---|---|---|
| Use moment-matched conditional covariance in the inequality-constraint score instead of the naive constraint approximation (ablation) | Conditional Diffusion · FitzHugh-Nagumo | tail-sample fidelity | Improved |
| Use moment-matched conditional covariance in the inequality-constraint score instead of the naive constraint approximation (ablation) | Conditional Diffusion · Lorenz attractor | event-conditional sampling | Improved |
| Apply the probabilistic conditional score to a nonlinear inequality defining neuron spikes | Diffusion model · FitzHugh-Nagumo | tail-sample fidelity | Improved |
| Apply the probabilistic conditional score to a nonlinear inequality separating trajectories that do not switch attractor arms | Diffusion model · Lorenz attractor | event-conditional sampling | Improved |
| Estimate event probability using likelihood-based estimators | Diffusion model · FitzHugh-Nagumo neuron spiking event | event-probability calibration | Hurt |
| Use an unconditional diffusion model to represent trajectory distributions | Diffusion model · Lorenz attractor, FitzHugh-Nagumo, and double pendulum | trajectory prediction relative error | Mixed |
| Use the learned diffusion model to represent predictive distributions | Diffusion model · Lorenz attractor | uncertainty calibration | Mixed |
Closest papers by meaning
- Predicting partially observable dynamical systems via diffusion models with a multiscale inference scheme (NeurIPS 2025 · similarity 0.89)
- Towards Non-Asymptotic Convergence for Diffusion-Based Generative Models (ICLR 2024 · similarity 0.88)
- Uncertainty Quantification for Large Language Diffusion Models (NeurIPS 2026 · similarity 0.88)
- Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics (ICLR 2025 · similarity 0.88)
- BayesDiff: Estimating Pixel-wise Uncertainty in Diffusion via Bayesian Inference (ICLR 2024 · similarity 0.88)
- Tracing the Roots: Leveraging Temporal Dynamics in Diffusion Trajectories for Origin Attribution (NeurIPS 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.
