danyang peng — accepted AI-conference papers

2026

Learning to Refine: Spectral-Decoupled Iterative Refinement Framework for Precipitation Nowcasting

Yunlong Zhou, Chen Zhao, danyang peng, Fanfan Ji, Xiaotong Yuan

ICML 2026poster

Accurate precipitation nowcasting is vital for disaster mitigation, but deep learning methods suffer a key trade-off: regression models produce over-smoothed, spectrally decaying predictions that blur convective details and violate turbulence power laws; diffusion models generate realistic yet unanc…

Cited by 0 Source ↗Scholar ↗

2026

LoPhyDA: Low-Rank Tensor and Physics Gradient Guided Diffusion for Atmospheric Data Assimilation

danyang peng, Yang Chen, Yunlong Zhou, Xiaotong Yuan

ICML 2026poster

Data Assimilation (DA) aims to integrate observations with model forecasts to estimate the state of dynamical systems. Despite the widespread application of diffusion-based assimilation methods, they remain constrained by the high dimensionality of atmospheric states and the reliance on imperfect st…

Cited by 0 Source ↗Scholar ↗