titleHiUni: Harmonizing Understanding and Generation in Shared Visual Space
authorsJiahao Hu, Jingdong Du, Yichen Zhang, Yi Song, Zhenyu Zhang, Jiaxin Niu, Ruihang Li, Xiangzhao Hao, Yiqian Zhang, Yong Huang, Zefeng Zhang, Zheyuan Gu, Shuohuan Wang
date2026-11-19
venuearXiv (preprint, not peer reviewed)
idarXiv:2611.26107
affiliationernie-research
upvotes23 (Hugging Face Daily Papers, 2026-11-21)
linksabs pdf hf code
cite
abstractUnified multimodal models aim to support visual understanding and generation within a single framework. However, prevailing approaches often separate their visual representations or backbone computation to accommodate the different requirements of these two tasks, which limits the degree of unification. In this paper, we present HiUni, which revisits this design choice and explores whether a pretrained vision language model can be directly extended into a unified model with extensive parameter sharing. Specifically, HiUni reuses the same language-aligned ViT and dense Transformer backbone for both capabilities, while confining task-specific computation to the output side. Yet such a design introduces two key challenges. Generation in pixel space is difficult to optimize, and semantic abstraction in pretrained ViT discards fine spatial details required for synthesis. To address these challenges, HiUni introduces a Pixel Diffusion Head tailored to direct pixel-space optimization and a lightweight HiGate pathway that provides fine-grained spatial details near the output stage. Extensive experiments show that HiUni achieves a similar average understanding score to the pretrained VLM while attaining competitive performance on image generation and editing. Controlled comparisons and analyses further demonstrate that a single dense backbone can effectively support both capabilities without task-specific backbone computation, highlighting a simpler path toward unified multimodal modeling.

bibtex

@misc{hu2026hiuni,
  title={HiUni: Harmonizing Understanding and Generation in Shared Visual Space},
  author={Jiahao Hu and Jingdong Du and Yichen Zhang and Yi Song and Zhenyu Zhang and Jiaxin Niu and Ruihang Li and Xiangzhao Hao and Yiqian Zhang and Yong Huang and Zefeng Zhang and Zheyuan Gu and Shuohuan Wang},
  year={2026},
  eprint={2611.26107},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2611.26107}
}