UniSpace: Unified Visual Representation and Scalable Multimodal Modeling

arXiv:2608.08676 · cs.CV, cs.AI · Submitted 2026-08-09 · Read on arXiv

Jinbo Yan, Limeng Qiao, Jie Qin, Junyan He, Feize Wu, Guanglu Wan

cs.CV, cs.AI

Submitted: 2026-08-09

Updated: 2026-08-11

Code: https://github.com/yjb6/UniSpace

Project page: https://yjb6.github.io/UniSpace

License: http://creativecommons.org/licenses/by/4.0/

The gist: Semantic vision encoders have become a central visual interface for multimodal understanding and semantic conditioning in image generation.

Terminology

Abstract

Semantic vision encoders have become a central visual interface for multimodal understanding and semantic conditioning in image generation. However, their final tokens discard fine-grained visual details, leading to poor pixel reconstruction and limiting their use in reconstruction-sensitive tasks such as image generation and editing. In this work, we ask whether understanding, generation, and editing can be modeled in a single visual representation space built from a pretrained semantic ViT. We show that the frozen Transformer blocks of a semantic ViT are not intrinsically unable to preserve visual details. Instead, the original patch parameterization drives the representation toward semantic abstraction, making fine-grained information difficult to recover from the final tokens. Based on this observation, we introduce Patch Reparameterization, which preserves the original semantic pathway while adding a reconstruction-aware patch embedding that provides fine-grained visual information to the same frozen ViT blocks. The resulting unified representation preserves multimodal understanding while enabling high-fidelity image reconstruction and a favorable reconstruction--generation trade-off. We further scale this representation into UniSpace, an 8B Mixture-of-Transformer-Experts model that performs understanding, generation, and editing in the same visual space without a separate VAE pathway. System-level evaluations demonstrate practical text-to-image generation and instruction-based image editing, showing that a reparameterized pretrained ViT can serve as a unified visual interface for scalable multimodal modeling.

Sources

Related papers