Show-o2: Improved Native Unified Multimodal Models
This work addresses the need for unified multimodal AI systems, but appears incremental as it builds upon existing methods like autoregressive modeling and flow matching.
The paper tackles the problem of building scalable multimodal models for understanding and generation across text, images, and videos, resulting in Show-o2 models that demonstrate versatility across diverse tasks.
This paper presents improved native unified multimodal models, \emph{i.e.,} Show-o2, that leverage autoregressive modeling and flow matching. Built upon a 3D causal variational autoencoder space, unified visual representations are constructed through a dual-path of spatial (-temporal) fusion, enabling scalability across image and video modalities while ensuring effective multimodal understanding and generation. Based on a language model, autoregressive modeling and flow matching are natively applied to the language head and flow head, respectively, to facilitate text token prediction and image/video generation. A two-stage training recipe is designed to effectively learn and scale to larger models. The resulting Show-o2 models demonstrate versatility in handling a wide range of multimodal understanding and generation tasks across diverse modalities, including text, images, and videos. Code and models are released at https://github.com/showlab/Show-o.