AI Paper Atlas

Research map / Probabilistic and scientific modeling / Differential equations and PDEs

P3D: Highly Scalable 3D Neural Surrogates for Physics Simulations with Global Context

Benjamin Holzschuh, Georg Kohl, Florian Redinger, Nils Thuerey

ICLR 2026 · Probabilistic and scientific modeling › Differential equations and PDEs · approach: Physics-informed operator learning

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 changedWhereMeasuredEffect
add transformer backbone on top of conv encoder/decoder (ablation)P3D-S · joint PDE, crop 64^3validation MSE x1e-3: 5.40 -> 2.77Improved
replace patch tokenizer with large convolutional encoder/decoder (ablation)P3D-S · joint PDE, crop 64^3validation MSE x1e-3: 3.78 -> 2.77Improved
add PDE type conditioning (ablation)P3D-S · joint PDE, crop 64^3validation MSE x1e-3: 2.94 -> 2.77Improved
enable shifted windows (ablation)P3D-S · joint PDEvalidation MSE x1e-3: 2.77 -> 2.84 (64^3)No clear effect
increase window size from 4 to 8 (ablation)P3D-S · joint PDEvalidation MSE x1e-3: 2.77 -> 2.90 (64^3), 1.35 -> 1.32 (128^3)No clear effect
scale model size from S to LP3D · joint PDE, crop 128^3nRMSE x1e-2: 3.33 -> 2.08Improved
train on larger cropsP3D-S · joint PDEnRMSE x1e-2: 6.27 (32^3) -> 3.76 (64^3) -> 3.33 (128^3)Improved
run full-domain inference via translation equivariance instead of independent cropsP3D-S · isotropic turbulence 512^3RMSE x1e-2: 1.90 -> 1.60Improved
add context network with finetuning to crop-pretrained model (ablation)P3D-L · turbulent channel flowmean L2 x1e-5: 5862.81 -> 23.6Improved
add region tokens in finetuning (ablation)P3D-L · turbulent channel flowmean L2 x1e-5: 4541 -> 23.6Improved
finetune decoder only for longer (20 to 500 epochs) (ablation)P3D-L · turbulent channel flowmean L2 x1e-5: 941 -> 16.8Improved
train P3D on crops and scale to full domain vs train on full domainP3D-L · turbulent channel flowmean L2 x1e-5: 3.02 (full domain) vs 16.8 (decoder-only finetune)Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy