Jingtao Li

Achieving state-of-the-art multimodal understanding at a fraction of the cost and data requirements challenges the notion that bigger models are always better.

HyperImageNet reveals that fine-grained land-cover classification can be significantly advanced with a dataset that combines raw imagery and detailed annotations.

Autoregressive image models can now compete with diffusion models in image quality and efficiency, thanks to a variable-length tokenization scheme that decouples compute from resolution.