Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914055657095168 |
|---|---|
| author | Lyu, Yuanhuiyi Wong, Chi Kit Liao, Chenfei Jiang, Lutao Zheng, Xu Lu, Zexin Zhang, Linfeng Hu, Xuming |
| author_facet | Lyu, Yuanhuiyi Wong, Chi Kit Liao, Chenfei Jiang, Lutao Zheng, Xu Lu, Zexin Zhang, Linfeng Hu, Xuming |
| contents | Recent works have made notable advancements in enhancing unified models for text-to-image generation through the Chain-of-Thought (CoT). However, these reasoning methods separate the processes of understanding and generation, which limits their ability to guide the reasoning of unified models in addressing the deficiencies of their generative capabilities. To this end, we propose a novel reasoning framework for unified models, Understanding-in-Generation (UiG), which harnesses the robust understanding capabilities of unified models to reinforce their performance in image generation. The core insight of our UiG is to integrate generative guidance by the strong understanding capabilities during the reasoning process, thereby mitigating the limitations of generative abilities. To achieve this, we introduce "Image Editing" as a bridge to infuse understanding into the generation process. Initially, we verify the generated image and incorporate the understanding of unified models into the editing instructions. Subsequently, we enhance the generated image step by step, gradually infusing the understanding into the generation process. Our UiG framework demonstrates a significant performance improvement in text-to-image generation over existing text-to-image reasoning methods, e.g., a 3.92% gain on the long prompt setting of the TIIF benchmark. The project code: https://github.com/QC-LY/UiG |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_18639 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generation Lyu, Yuanhuiyi Wong, Chi Kit Liao, Chenfei Jiang, Lutao Zheng, Xu Lu, Zexin Zhang, Linfeng Hu, Xuming Computer Vision and Pattern Recognition Recent works have made notable advancements in enhancing unified models for text-to-image generation through the Chain-of-Thought (CoT). However, these reasoning methods separate the processes of understanding and generation, which limits their ability to guide the reasoning of unified models in addressing the deficiencies of their generative capabilities. To this end, we propose a novel reasoning framework for unified models, Understanding-in-Generation (UiG), which harnesses the robust understanding capabilities of unified models to reinforce their performance in image generation. The core insight of our UiG is to integrate generative guidance by the strong understanding capabilities during the reasoning process, thereby mitigating the limitations of generative abilities. To achieve this, we introduce "Image Editing" as a bridge to infuse understanding into the generation process. Initially, we verify the generated image and incorporate the understanding of unified models into the editing instructions. Subsequently, we enhance the generated image step by step, gradually infusing the understanding into the generation process. Our UiG framework demonstrates a significant performance improvement in text-to-image generation over existing text-to-image reasoning methods, e.g., a 3.92% gain on the long prompt setting of the TIIF benchmark. The project code: https://github.com/QC-LY/UiG |
| title | Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.18639 |