Wei-Yan Wang

Robotic world models can now learn from diverse, yet confident, alternative futures, achieving state-of-the-art prediction quality and extending predictive horizons.

Robot video world models can be significantly improved by distilling a multimodal reward function and stabilizing long-horizon inference, leading to better instruction following and manipulation accuracy.

Model-based RL can leverage non-differentiable domain metrics to drastically improve spatiotemporal forecasting, especially for capturing extreme events.