A transformer with prediction-correction and hierarchical super-token merging unifies simulation of six physical dynamics categories on Lagrangian particles and generalizes to unseen conditions.
arXiv preprint arXiv:2407.03925 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
PAINET proposes an SE(3)-equivariant transformer with physics-inspired attention from energy minimization for 3D dynamics modeling, reporting 4.7-41.5% error reductions on human motion, molecular, and protein benchmarks.
citing papers explorer
-
WorldParticle: Unified World Simulation of Lagrangian Particle Dynamics via Transformer
A transformer with prediction-correction and hierarchical super-token merging unifies simulation of six physical dynamics categories on Lagrangian particles and generalizes to unseen conditions.
-
PAINET: A Principled Efficient Transformer for 3D Dynamics Modeling
PAINET proposes an SE(3)-equivariant transformer with physics-inspired attention from energy minimization for 3D dynamics modeling, reporting 4.7-41.5% error reductions on human motion, molecular, and protein benchmarks.