Xuanze Zhang

Imagine an interpreter in a huge hall who hears not only the person next to them but also a whisper in the far corner, and from those fragments reconstructs the entire conversation. That's how the CAS-Canglong neural network works: it picks up connections between pressure, wind, and heat across the whole planet and uses them to predict what the ocean will look like three months from now.

Imagine a translator sitting in on a conversation between two parties: the ocean and the atmosphere. Usually, the translator conveys the meaning with a six-month delay, and in spring it loses the thread entirely. The new model, called CTEFNet, translates up to two years ahead and, on top of that, explains which remark turned the tide.