AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

UniPixie: Unified and Probabilistic 3D Physics Learning via Flow Matching

Qilin Huang, Quynh Anh Huynh, Long Le, Chen Wang, Chuhao Chen, Ryan Lucas, Eric Eaton, Lingjie Liu

CVPR 2026 · Probabilistic and scientific modeling › Dynamical systems · approach: Physics-informed dynamics learning · 1 citations (OpenAlex)

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
replace 3D U-Net diffusion with flow matching transformer (ablation)UniPixie · PixieMultiVerse testlog E MSE: 0.0410 -> 0.0091Improved
replace 3D U-Net diffusion with flow matching transformer (ablation)UniPixie · PixieMultiVerse testlog rho MSE: 0.1464 -> 0.0194Improved
replace 3D U-Net diffusion with flow matching transformer (ablation)UniPixie · PixieMultiVerse testnu MSE: 0.4604 -> 0.0240Improved
replace 3D U-Net diffusion with flow matching transformer (ablation)UniPixie · PixieMultiVerse testmaterial accuracy: 96.3% -> 93.9%Hurt
train a generative model over softest-to-stiffest spectrum instead of deterministic midpoint regressionPIXIE · PixieMultiVerse testlog E MSE: 0.0205 -> 0.0091Improved
generative spectrum training instead of deterministic regressionPIXIE · PixieMultiVerse testmaterial accuracy: 97.3% -> 93.9%Hurt
single unified model instead of specialist for MPMPIXIE · MPM, alpha=0.0PSNR: 23.16 -> 29.25Improved
single unified model instead of specialist for MPMPIXIE · MPM, alpha=0.5SSIM: 0.9225 -> 0.9198Hurt
single unified model instead of Vid2Sim (full) for LBSVid2Sim (full) · LBS, alpha=0.0PSNR: 28.30 -> 33.83Improved
single unified model instead of Vid2Sim (full) for LBSVid2Sim (full) · LBS, alpha=0.5PSNR: 36.94 -> 36.81No clear effect
single unified model instead of tuned Spring-GausSpring-Gaus (tuned) · Spring-Mass, alpha=0.0PSNR: 30.53 -> 33.88Improved
feed-forward prediction instead of test-time optimizationSpring-Gaus · Spring-Massruntime (s): 4375.30 -> 21.64Improved

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