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.
Title resolution pending
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
fields
cs.GR 4years
2026 4verdicts
UNVERDICTED 4representative citing papers
AGIPC achieves up to 3x speedup in GPU IPC by performing adaptive algebraic coarsening inside the solver while keeping results visually indistinguishable from the fine-scale version.
YASPS is a symbolic differentiable framework that uses JOIN and UNION operators to enable extensible high-performance IPC simulation on GPUs with automatic sparsity derivation and JIT kernel compilation.
Proposes a Koopman-surrogate framework that turns cyclic animation synthesis into a structured quadratic program solved via KKT system under temporal periodicity.
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.
-
AGIPC: Adaptive In-Solve Algebraic Coarsening for GPU IPC
AGIPC achieves up to 3x speedup in GPU IPC by performing adaptive algebraic coarsening inside the solver while keeping results visually indistinguishable from the fine-scale version.
-
YASPS: A Symbolic Framework for Extensible, High-Performance IPC Simulation
YASPS is a symbolic differentiable framework that uses JOIN and UNION operators to enable extensible high-performance IPC simulation on GPUs with automatic sparsity derivation and JIT kernel compilation.
-
Closing Trajectories: Equation-Free Cyclic Animation via Koopman Surrogates
Proposes a Koopman-surrogate framework that turns cyclic animation synthesis into a structured quadratic program solved via KKT system under temporal periodicity.