Research map / Probabilistic and scientific modeling / Molecules and proteins
Generative Neural Operators through Diffusion Last Layer
Sungwon Park, Anthony Zhou, Hongjoong Kim, Amir Barati Farimani
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 |
|---|---|---|---|
| add diffusion last layer to FNO for stochastic prediction | FNO · Stochastic Burgers | Energy distance: 6.491 -> 1.285 | Improved |
| add diffusion last layer to FNO for stochastic prediction | FNO · Stochastic Darcy flow | Energy distance: 1.463 -> 0.227 | Improved |
| add diffusion last layer to FNO | FNO · Stochastic Burgers | NRMSE of mean: 0.146 -> 0.252 | Hurt |
| add diffusion last layer to FNO for rollout | FNO · Kuramoto-Sivashinsky rollout | NRMSE: 0.404 -> 0.343 | Improved |
| add diffusion last layer to FNO for rollout | FNO · Kolmogorov flow rollout | NRMSE: 0.528 -> 0.426 | Improved |
| use DLL instead of pixel-space diffusion (DM) | DM · Kolmogorov flow rollout | NRMSE: 0.369 -> 0.426 | Hurt |
| use DLL instead of conventional latent diffusion (LDM) | LDM · KS rollout | NRMSE: 0.576 -> 0.343 | Improved |
| use DLL instead of PNO (Gaussian probabilistic operator) | PNO · Stochastic Burgers | Energy distance: 1.766 -> 1.285 | Improved |
| use operator encoder (input-dependent basis) instead of autoencoder (ablation) | latent representation · deterministic benchmarks (KS, Kolmogorov) | reconstruction NRMSE: 7.75 × 10−4 -> 2.45 × 10−4 (KS) | Improved |
| use operator encoder instead of autoencoder (ablation) | latent representation · stochastic benchmarks (Burgers, Darcy) | reconstruction NRMSE: 1.98 × 10−3 -> 4.13 × 10−2 (Burgers) | Hurt |
| add MC dropout to FNO (FNO-d) | FNO · Stochastic Burgers | Energy distance: 6.491 -> 6.075 | No clear effect |
Closest papers by meaning
- Light-Weight Diffusion Multiplier and Uncertainty Quantification for Fourier Neural Operators (NeurIPS 2025 · similarity 0.89)
- Learning to Solve Generative ODEs Beyond the Linear Span (NeurIPS 2026 · similarity 0.86)
- Using Uncertainty Quantification to Characterize and Improve Out-of-Domain Learning for PDEs (ICML 2024 · similarity 0.86)
- MgNO: Efficient Parameterization of Linear Operators via Multigrid (ICLR 2024 · similarity 0.86)
- Linearization Turns Neural Operators into Function-Valued Gaussian Processes (ICML 2025 · similarity 0.86)
- Optimal deep learning of holomorphic operators between Banach spaces (NeurIPS 2024 · similarity 0.86)
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.
