topic

new physics

New physics refers to theoretical frameworks, hypothetical particles, and physical phenomena that extend beyond or cannot be explained by the standard model of particle physics. It addresses foundational gaps in contemporary scientific understanding, such as the nature of dark matter and dark energy, the origin of neutrino masses, and the integration of gravity with quantum mechanics. In high-energy physics and scientific computing, the pursuit of new physics involves large-scale numerical simulations, complex data-analysis pipelines, and machine learning algorithms designed to detect subtle deviations and rare event signatures in collider experiments and astrophysical observations.

1 item

Transolver: A Fast Transformer Solver for PDEs on General Geometries

Haixu Wu, Huakun Luo, Haowen Wang, Jianmin Wang, Mingsheng Long

OrganizationsTsinghua University

Why you should read this

Develops Transolver, a linear-complexity Transformer architecture for solving partial differential equations on complex geometries by adaptively grouping mesh points into physics-aware slices, achieving a 22% relative performance gain across standard benchmarks and scaling to large-scale industrial aerodynamic simulations.

Transformers have empowered many milestones across various fields and have recently been applied to solve partial differential equations (PDEs). However, since PDEs are typically discretized into large-scale meshes with complex geometries, it is challenging for Transformers to capture intricate physical correlations directly from massive individual points. Going beyond superficial and unwieldy meshes, we present Transolver based on a more foundational idea, which is learning intrinsic physical states hidden behind discretized geometries. Specifically, we propose a new Physics-Attention to adaptively split the discretized domain into a series of learnable slices of flexible shapes, where mesh points under similar physical states will be ascribed to the same slice. By calculating attention to physics-aware tokens encoded from slices, Transovler can effectively capture intricate physical correlations under complex geometrics, which also empowers the solver with endogenetic geometry-general modeling capacity and can be efficiently computed in linear complexity. Transolver achieves consistent state-of-the-art with 22% relative gain across six standard benchmarks and also excels in large-scale industrial simulations, including car and airfoil designs. Code is available at this https URL.

Source

https://arxiv.org/pdf/2402.02366

Added

2026-09-28