HiUni model release
The story so far
Hugging Face released the HiUni model, extending pre-trained visual language models into unified models through shared language-aligned ViT and dense Transformer backbones, achieving visual understanding and generation capabilities. The model uses Pixel Diffusion Head for pixel space optimization and introduces a lightweight HiGate path to provide fine spatial details, achieving average scores similar to pre-trained VLMs in understanding tasks, while achieving competitive performance in image generation and editing.
AI-written from the reports · Updated 1 hour ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersHiUni: Harmonizing Understanding and Generation in Shared Visual Space
HiUni extends pre-trained visual language models into unified models by sharing language-aligned ViT and dense Transformer backbones, achieving visual understanding and generation capabilities. The model adopts Pixel Diffusion Head for pixel space optimization and introduces a lightweight HiGate path to provide fine spatial details, achieving average scores similar to pre-trained VLMs in understanding tasks, while also achieving competitive performance in image generation and editing.
Heat over time
Not enough data yet to draw a trend.