Studying Image Tokenizers as Visual Languages in Unified Multimodal Models
A controlled pure-autoregressive testbed is built and track task-specific validation losses during multimodal continual pretraining across text, image, text-to-image (T2I), and image-to-text (I2T) prediction, showing that better reconstruction does not necessarily yield lower task-specific losses or stronger downstream...
Siting Li, Zheng-Yang Wang, S. Du et al.
· 0 citations