Zesheng Liu

COGENT achieves long-term stability in physical forecasting by enabling predictions at any future time, not just fixed intervals.

Stop letting your climate models drift: a novel graph neural network architecture predicts geophysical system evolution across multiple time horizons with improved long-range accuracy and stability.

Synthesizing complete ice-layer data from sparse radar traces is now possible by conditioning on climate model outputs, enabling more accurate analysis of ice dynamics and snow accumulation.