Walrus: A Cross-Domain Foundation Model for Continuum Dynamics
This work addresses the problem of data heterogeneity and unstable dynamics in physical simulation for researchers in fields like astrophysics and fluid dynamics, representing a strong but incremental advance over existing methods.
The authors tackled the challenge of building a foundation model for physical simulation by developing Walrus, a transformer-based model for continuum dynamics, which outperforms prior foundation models on both short and long-term predictions across diverse scenarios.
Foundation models have transformed machine learning for language and vision, but achieving comparable impact in physical simulation remains a challenge. Data heterogeneity and unstable long-term dynamics inhibit learning from sufficiently diverse dynamics, while varying resolutions and dimensionalities challenge efficient training on modern hardware. Through empirical and theoretical analysis, we incorporate new approaches to mitigate these obstacles, including a harmonic-analysis-based stabilization method, load-balanced distributed 2D and 3D training strategies, and compute-adaptive tokenization. Using these tools, we develop Walrus, a transformer-based foundation model developed primarily for fluid-like continuum dynamics. Walrus is pretrained on nineteen diverse scenarios spanning astrophysics, geoscience, rheology, plasma physics, acoustics, and classical fluids. Experiments show that Walrus outperforms prior foundation models on both short and long term prediction horizons on downstream tasks and across the breadth of pretraining data, while ablation studies confirm the value of our contributions to forecast stability, training throughput, and transfer performance over conventional approaches. Code and weights are released for community use.