Fetching the paper…

VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters · Around