Walrus 物理基础模型从模拟到实验室湍流的零样本迁移
Emergent Transfer of a Physics Foundation Model from Simulation to Laboratory Turbulence
在三个以内 DNS 实现上微调的连续动力学基础模型 Walrus，零样本应用于滑动屏障实验室数据时，从类 DNS 区间进入实验观测的混合增长率区间（α 约 0.06-0.07，约为理想 DNS 约 0.02 的三倍），且从未见过任何实验样本。该结果独立支持初始条件在长期存在的模拟-实验 α 差距中起关键作用。模型还零样本泛化到训练中未出现的稳定分层浮力情形，正确减缓混合层增长。
Authors:Payel Mukhopadhyay, Stefan S. Nixon, Romain Watteaux, Michael McCabe, Alberto Bietti, Kyunghyun Cho, Cristiana Diaconu, Irina Espejo Morales, David Fouhey, Siavash Golkar, Tom Hehir, Shirley Ho, Jake Kovalic, Geraud Krawezik, Francois Lanusse, Tanya Marwah, Rudy Morel, Mariel Pettee, Helen Qu, Jeff Shen, Hadi Sotoudeh, Stuart B. Dalziel, Miles Cranmer
Abstract:Whether physics foundation models can be usefully deployed on laboratory experiments remains an open question for scientific machine learning (ML). We test this question on the Rayleigh-Taylor instability (RTI), a ubiquitous and demanding fluid instability seen from tabletop flows to supernova explosions, in which small perturbations at a density interface grow into chaotic, multiscale mixing as a lighter fluid accelerates into a heavier one. Standard ML models struggle with RTI, and despite over a century of theoretical, numerical, and experimental work, it carries an unresolved discrepancy between simulation and experiment: the late-time mixing growth rate, $\alpha$, measured in most laboratory experiments ($\sim$ 0.06-0.07), is roughly three times the value from idealized direct numerical simulations (DNS, $\sim$ 0.02). The gap's origin remains debated. These properties make RTI a stringent test for a question that matters well beyond RTI: can foundation models trained only on simulations generalise to sparse, messy, and noisy laboratory settings? We finetune Walrus, a foundation model for continuum dynamics, on three or fewer DNS realizations and recover key RTI physics over long rollouts. Applied zero-shot to sliding-barrier laboratory data, the finetuned model leaves the DNS-like regime and enters the observed growth band, having never seen a single experimental sample. These results provide independent, data-driven evidence that initial conditions play a crucial role in the longstanding sim-experiment gap in $\alpha$. The model also generalises zero-shot to stable stratification, a buoyancy regime absent from training, correctly slowing mixing-layer growth. Together, our results show that foundation models can generalise well beyond their training data, predicting laboratory behavior and unseen physical regimes, opening new ways to probe longstanding simulation-experiment gaps.
| Subjects: | Fluid Dynamics (physics.flu-dyn); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2606.01470 [physics.flu-dyn] |
| (or arXiv:2606.01470v2 [physics.flu-dyn] for this version) | |
| https://doi.org/10.48550/arXiv.2606.01470 arXiv-issued DOI via DataCite |
Submission history
From: Payel Mukhopadhyay [view email]
[v1]
Sun, 31 May 2026 22:07:38 UTC (11,527 KB)
[v2]
Wed, 4 Nov 2026 02:58:03 UTC (11,527 KB)
来源：arXiv cs.AI · arxiv.org