mlsanityData to 14 Oct 2026

LEIA: Learned Environment for Interactive Architected Materials

arXiv 2605.28368 ↗First posted 27 May 2026cs.LG cond-mat.mtrl-sci physics.app-phcited by 0

On the map: World models → Latent predictive world models

Abstract (arXiv)

World models have enabled interactive exploration of game environments and robotic manipulation, but physical engineering remains beyond their reach: real materials exhibit nonlinear constitutive laws, carry history-dependent internal state, undergo inertial dynamics, and may possess hierarchical structures spanning multiple length scales. We present LEIA (Learned Environment for Interactive Architected materials), a world model that lets engineers apply boundary conditions step by step and observe the resulting deformation and stress fields in real time. LEIA handles large three-dimensional unstructured meshes and generates autoregressive responses to user-specified loading. We introduce MicroPlate, a benchmark of architected plates spanning two regimes of microstructure modeling: architected lattices that resolve microstructure explicitly through three-dimensional geometry, and a homogeneous plate where microstructural change is modeled implicitly through internal degrees of freedom. MicroPlate is used to assess LEIA alongside four baseline methods across both regimes. Finally, we demonstrate that LEIA enables efficient candidate generation and ranking for fast surrogate-guided search for de novo designs of architected materials, with stress-accurate candidate ranking validated by finite element ground truth.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Latent predictive world models, one of 11 directions in World models. Builds on 31 of its 69 references that are in the map. First posted on arXiv 27 May 2026.

Builds on

31 of 69 references.

References · 31

  1. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  2. 2023Mastering Diverse Domains through World Modelscited by 1,253
  3. 2020Learning Mesh-Based Simulation with Graph Networkscited by 868
  4. 2018Recurrent World Models Facilitate Policy Evolutioncited by 785
  5. 2024Genie: Generative Interactive Environmentscited by 747
  6. 2021Perceiver IO: A General Architecture for Structured Inputs & Outputscited by 629
  7. 2023Learning Interactive Real-World Simulatorscited by 458
  8. 2022Message Passing Neural PDE Solverscited by 386
  9. 2017Sobolev Training for Neural Networkscited by 143
  10. 2025Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometriescited by 97
  11. 2024Physics-Informed Diffusion Modelscited by 90
  12. 2023Solving High-Dimensional PDEs with Latent Spectral Modelscited by 89
19 more
  1. 2024Latent Neural Operator for Solving Forward and Inverse PDE Problemscited by 80
  2. 2024AROMA: Preserving Spatial Structure for Latent PDE Modeling with Local Neural Fieldscited by 39
  3. 2024Universal Physics Transformers: A Framework For Efficiently Scaling Neural Operatorscited by 27
  4. 2024Derivative-enhanced Deep Operator Networkcited by 19
  5. 2024SineNet: Learning Temporal Dynamics in Time-Dependent Partial Differential Equationscited by 14
  6. 2025VoMP: Predicting Volumetric Mechanical Property Fieldscited by 13
  7. 2025Automotive Crash Dynamics Modeling Accelerated with Machine Learningcited by 10
  8. 2026MeshGraphNet-Transformer: Scalable Mesh-based Learned Simulation for Solid Mechanicscited by 10
  9. 2026GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Trainingcited by 10
  10. 2025Geometry-Informed Neural Operator Transformercited by 8
  11. 2025P3D: Scalable Neural Surrogates for High-Resolution 3D Physics Simulations with Global Contextcited by 7
  12. 2025Derivative-Informed Fourier Neural Operator: Universal Approximation and Applications to PDE-Constrained Optimizationcited by 7
  13. 2025Improving Long-Range Interactions in Graph Neural Simulators via Hamiltonian Dynamicscited by 6
  14. 2025Learning Memory and Material Dependent Constitutive Lawscited by 5
  15. 2024A physics-encoded Fourier neural operator approach for surrogate modeling of divergence-free stress fields in solidscited by 4
  16. 2025Sobolev acceleration for neural networkscited by 3
  17. 2025Adaptive Mesh-Quantization for Neural PDE Solverscited by 2
  18. 2025Automating modeling in mechanics: LLMs as designers of physics-constrained neural networks for constitutive modeling of materialscited by 2
  19. 2025A comprehensive comparison of neural operators for 3D industry-scale engineering designscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Generative AI for material design: A mechanics perspective from burgers to matter
  2. 2026PhysCoRe: Physics-Corrected Residual World Models for Material-Aware Deformable Dynamics
  3. 2026UniPixie: Unified and Probabilistic 3D Physics Learning via Flow Matching
  4. 2025MIND: Microstructure INverse Design with Generative Hybrid Neural Representation
  5. 2026DiffGA: Steering Frozen Diffusion Models for Materials Design via Evolutionary Search
  6. 2024Physics3D: Learning Physical Properties of 3D Gaussians via Video Diffusion
  7. 2026Co-PiLOT: Constrained Physics-Informed Latent Optimization for Target-Driven Inverse Design
  8. 2024Phys4DGen: Physics-Compliant 4D Generation with Multi-Material Composition Perception