Vision Bridge Transformer at Scale
This work addresses efficient data-to-data translation for image editing and video generation, representing an incremental advancement in scaling Bridge Models.
The paper tackles the problem of conditional generation by scaling Vision Bridge Transformer (ViBT) to 20B and 1.3B parameters, demonstrating effectiveness for image and video translation tasks.
We introduce Vision Bridge Transformer (ViBT), a large-scale instantiation of Brownian Bridge Models designed for conditional generation. Unlike traditional diffusion models that transform noise into data, Bridge Models directly model the trajectory between inputs and outputs, creating an efficient data-to-data translation paradigm. By scaling these models to 20B and 1.3B parameters, we demonstrate their effectiveness for image and video translation tasks. To support this scale, we adopt a Transformer architecture and propose a variance-stabilized velocity-matching objective for robust training. Together, these advances highlight the power of scaling Bridge Models for instruction-based image editing and complex video translation.