用合成动态数据预训练几何模型，提升物理模拟效率与泛化能力。
GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training
- 基于几何结构生成合成动态数据，实现无物理标签的自监督学习。
- 在百万级样本上预训练，使标注数据需求减少20%-60%，收敛速度提升2倍。
- 适用于汽车、飞机、船舶流体及碰撞模拟，适合工业级物理仿真场景。
神经网络模拟器有望成为物理模拟的高效替代方案，但其规模化受限于高保真训练数据的生成成本。在现成几何体上进行预训练是一种自然选择，但存在根本缺陷：仅依赖静态几何的监督会忽略动力学信息，导致物理任务中出现负迁移。我们提出GeoPT，一种基于提升几何预训练的通用物理模拟统一模型。核心思想是通过合成动态数据增强几何结构，实现无需物理标签的动力学感知自监督。该模型在超过一百万样本上预训练，显著提升涵盖汽车、飞机、船舶流体力学和碰撞模拟的工业级基准表现，使标注数据需求降低20%-60%，收敛速度加快2倍。结果表明，合成动态提升有效弥合了几何与物理之间的鸿沟，为神经模拟乃至更广泛领域提供了可扩展路径。代码已开源。
原文摘要 · Abstract (English)
Neural simulators promise efficient surrogates for physics simulation, but scaling them is bottlenecked by the prohibitive cost of generating high-fidelity training data. Pre-training on abundant off-the-shelf geometries offers a natural alternative, yet faces a fundamental gap: supervision on static geometry alone ignores dynamics and can lead to negative transfer on physics tasks. We present GeoPT, a unified pre-trained model for general physics simulation based on lifted geometric pre-training. The core idea is to augment geometry with synthetic dynamics, enabling dynamics-aware self-supervision without physics labels. Pre-trained on over one million samples, GeoPT consistently improves industrial-fidelity benchmarks spanning fluid mechanics for cars, aircraft, and ships, and solid mechanics in crash simulation, reducing labeled data requirements by 20-60% and accelerating convergence by 2$\times$. These results show that lifting with synthetic dynamics bridges the geometry-physics gap, unlocking a scalable path for neural simulation and potentially beyond. Code is available at https://github.com/Physics-Scaling/GeoPT.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。