arXiv：cs.LG· Yixuan Jia, Siyi Chen, Yida Pan, Xiao Li, Lianghe Shi, Chanyong Jung, Haijie Yuan, Ismail Alkhouri, Yue Cynthia Wu, Saiprasad Ravishankar, Jeffrey A Fessler, Qing Qu·· 4 小时前AI 评分34
ForcingDAS：基于 Diffusion Forcing 的统一鲁棒数据同化框架
ForcingDAS: Unified and Robust Data Assimilation via Diffusion Forcing
AI 导读
ForcingDAS 是一个基于 Diffusion Forcing 的统一数据同化框架，为每帧分配独立噪声水平、学习联合轨迹先验而非逐帧转移，从而捕捉长程时序依赖并减少误差累积。同一训练模型仅通过推理调度即可覆盖滤波到平滑全谱段，支持临近预报、固定滞后平滑与批量再分析，无需重新训练。在 2D Navier-Stokes 涡度、降水临近预报和全球大气状态估计上，单一模型媲美或超越各专门基线。
正文
Authors:Yixuan Jia, Siyi Chen, Yida Pan, Xiao Li, Lianghe Shi, Chanyong Jung, Haijie Yuan, Ismail Alkhouri, Yue Cynthia Wu, Saiprasad Ravishankar, Jeffrey A Fessler, Qing Qu
Abstract:Data assimilation (DA) estimates the state of an evolving dynamical system from noisy, partial observations, and is widely used in scientific simulation as well as weather and climate science. In practice, filtering methods rely on frame-to-frame transition models. However, these models are fragile when observations are non-Markovian (when they form only a partial slice of a higher-dimensional latent state as in real-world weather data): they tend to accumulate errors over long horizons. At the same time, learned DA methods typically commit to a single regime, either filtering (nowcasting, real-time forecasting) or smoothing (retrospective reanalysis), which splits what should be a shared prior across application-specific pipelines. To address both issues, we introduce ForcingDAS, a unified and robust DA framework. Built on Diffusion Forcing with an independent noise level assigned to each frame, ForcingDAS learns a joint-trajectory prior instead of frame-to-frame transitions. This allows it to capture long-horizon temporal dependencies and reduce error accumulation. In addition, the same trained model spans the full filtering to smoothing spectrum at inference time. Specifically, nowcasting, fixed-lag smoothing, and batch reanalysis are selected through the inference schedule alone, without retraining. We evaluate ForcingDAS on 2D Navier-Stokes vorticity, precipitation nowcasting, and global atmospheric state estimation. Across all settings, a single model is competitive with or outperforms both learned and classical baselines that are specialized for individual regimes, with the largest gains observed on real-world weather benchmarks.
| Subjects: | Image and Video Processing (eess.IV); Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.14285 [eess.IV] |
| (or arXiv:2605.14285v3 [eess.IV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.14285 arXiv-issued DOI via DataCite |
Submission history
From: Yixuan Jia [view email]
[v1]
Thu, 14 May 2026 02:34:33 UTC (36,211 KB)
[v2]
Fri, 5 Jun 2026 19:20:34 UTC (36,997 KB)
[v3]
Mon, 5 Oct 2026 20:45:37 UTC (14,621 KB)
来源：arXiv：cs.LG · arxiv.org