TUNA: Taming Unified Visual Representations for Native Unified Multimodal Models
By: Zhiheng Liu , Weiming Ren , Haozhe Liu and more
Potential Business Impact:
Lets computers understand and create pictures and videos.
Unified multimodal models (UMMs) aim to jointly perform multimodal understanding and generation within a single framework. We present TUNA, a native UMM that builds a unified continuous visual representation by cascading a VAE encoder with a representation encoder. This unified representation space allows end-to-end processing of images and videos for both understanding and generation tasks. Compared to prior UMMs with decoupled representations, TUNA's unified visual space avoids representation format mismatches introduced by separate encoders, outperforming decoupled alternatives in both understanding and generation. Moreover, we observe that stronger pretrained representation encoders consistently yield better performance across all multimodal tasks, highlighting the importance of the representation encoder. Finally, in this unified setting, jointly training on both understanding and generation data allows the two tasks to benefit from each other rather than interfere. Our extensive experiments on multimodal understanding and generation benchmarks show that TUNA achieves state-of-the-art results in image and video understanding, image and video generation, and image editing, demonstrating the effectiveness and scalability of its unified representation design.
Similar Papers
UniModel: A Visual-Only Framework for Unified Multimodal Understanding and Generation
CV and Pattern Recognition
Makes computers see and create pictures from words.
EMMA: Efficient Multimodal Understanding, Generation, and Editing with a Unified Architecture
CV and Pattern Recognition
Makes computers understand and create images and text.
Can Understanding and Generation Truly Benefit Together -- or Just Coexist?
CV and Pattern Recognition
Makes computers draw pictures from descriptions.