AI 中文总结
该研究针对深度生成模型忽略物理结构、物理模型不完善的问题，提出ClimPhyDM框架，在ERA5基准上优于对比模型，可在单12GB消费级GPU训练，提升了天气预报的长时稳定性。
AI 中文摘要
流匹配和扩散模型等深度生成模型在学习复杂动力系统方面展现出潜力，但通常作为黑箱模型忽略底层物理结构；而由偏微分方程控制的基于物理的模型常因缺失源项或参数化不确定而不完善。我们提出气候物理动态匹配（ClimPhyDM），这是一种用于天气预报的无模拟变分动力学驱动框架，在变分框架中将平流型物理先验与数据驱动组件结合，以捕捉未解析大气动力学的随机性和多模态。在ERA5基准数据集上，针对每小时（42小时）和每月（5个月）分辨率，ClimPhyDM的性能优于ClimODE和GB-DM，在长时程内保持更低误差，表明其时间稳定性和抗误差累积能力提升；同时，其无模拟范式还支持在单个普通12GB消费级GPU上进行训练。
英文摘要
Deep generative models such as flow matching and diffusion models have shown potential for learning complex dynamical systems, but typically act as black boxes that neglect underlying physical structure, while physics-based models governed by partial differential equations are often incomplete due to missing source terms, or uncertain parametrisations. We present Climate Physics Dynamic Matching (ClimPhyDM), a variational simulation-free dynamics informed framework for weather forecasting that combines an advection-type physics prior with data-driven components in a variational framework. % to capture the stochasticity and multi-modality of unresolved atmospheric dynamics. On the ERA5 benchmark at hourly (42-hour) and monthly (5-month) resolutions, ClimPhyDM outperforms ClimODE, and GB-DM, keeping the lower error at extended horizon, indicating improved temporal stability and resistance to error accumulation, while its simulation-free paradigm also enables training on a single modest 12 GB consumer GPU.
