跳到正文
arXiv cs.AI· Yilong Dai, Yiming Sun, Shengyu Chen, Jiaqi Gong, Xiaowei Jia, Runlong Yu·· 1 小时前AI 评分41

PhysDial:将预训练视频模型适配到变化物理条件下的 PDE 模拟

PhysDial: Adapting Pretrained Video Models for PDE Simulation under Changing Physical Conditions

AI 导读

PhysDial 将预训练视频扩散 Transformer 适配到八类二维 PDE 家族,通过系数随噪声水平输入、条件场作参考帧、方程文本作提示词实现条件控制。

正文 · 原文

View PDF HTML (experimental)

Abstract:Simulating how physical systems respond to changing conditions is central to control, design, and scenario studies. To our knowledge, prior PDE foundation models have not been trained or evaluated on changes within trajectories. Our counterfactual test measures how far predictions move toward a changed solver branch from the same state. Models trained on fixed conditions follow changes only partly. A PDE foundation model given a condition input after pretraining barely responds. PhysDial adapts a pretrained video diffusion transformer to eight two-dimensional PDE families. Coefficients enter with the noise level, condition fields as reference frames, and equation text as a prompt. In Kolmogorov flow, switches within trajectories account for about 5% of the training samples. At matched compute, they halve the error after a change in forcing wavenumber or amplitude. A few trajectories at new amplitudes more than halve extrapolation error. Without physics pretraining, PhysDial matches or exceeds PDE foundation models under training conditions on two turbulent flows. At step 64 on the Kolmogorov test split, PhysDial has 1.07 times the solver's small-scale energy and Walrus 7.1 times. Our experiments suggest that a learned simulator uses a condition only as far as its training data make that condition necessary for prediction. Learning from trajectories that diverge under intervention can help scientific simulators predict what would happen if a system's forcing or parameters changed.
Comments: Code: this https URL
Subjects: Fluid Dynamics (physics.flu-dyn); Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.23498 [physics.flu-dyn]
  (or arXiv:2611.23498v1 [physics.flu-dyn] for this version)
  https://doi.org/10.48550/arXiv.2611.23498

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yilong Dai [view email]
[v1] Tue, 17 Nov 2026 20:49:17 UTC (11,383 KB)

来源:arXiv cs.AI · arxiv.org