Enhancing Spatial Understanding in Image Generation via Reward Modeling
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908856267833344 |
|---|---|
| author | Tang, Zhenyu Feng, Chaoran Deng, Yufan Wu, Jie Li, Xiaojie Wang, Rui Chen, Yunpeng Zhou, Daquan |
| author_facet | Tang, Zhenyu Feng, Chaoran Deng, Yufan Wu, Jie Li, Xiaojie Wang, Rui Chen, Yunpeng Zhou, Daquan |
| contents | Recent progress in text-to-image generation has greatly advanced visual fidelity and creativity, but it has also imposed higher demands on prompt complexity-particularly in encoding intricate spatial relationships. In such cases, achieving satisfactory results often requires multiple sampling attempts. To address this challenge, we introduce a novel method that strengthens the spatial understanding of current image generation models. We first construct the SpatialReward-Dataset with over 80k preference pairs. Building on this dataset, we build SpatialScore, a reward model designed to evaluate the accuracy of spatial relationships in text-to-image generation, achieving performance that even surpasses leading proprietary models on spatial evaluation. We further demonstrate that this reward model effectively enables online reinforcement learning for the complex spatial generation. Extensive experiments across multiple benchmarks show that our specialized reward model yields significant and consistent gains in spatial understanding for image generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_24233 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Enhancing Spatial Understanding in Image Generation via Reward Modeling Tang, Zhenyu Feng, Chaoran Deng, Yufan Wu, Jie Li, Xiaojie Wang, Rui Chen, Yunpeng Zhou, Daquan Computer Vision and Pattern Recognition Recent progress in text-to-image generation has greatly advanced visual fidelity and creativity, but it has also imposed higher demands on prompt complexity-particularly in encoding intricate spatial relationships. In such cases, achieving satisfactory results often requires multiple sampling attempts. To address this challenge, we introduce a novel method that strengthens the spatial understanding of current image generation models. We first construct the SpatialReward-Dataset with over 80k preference pairs. Building on this dataset, we build SpatialScore, a reward model designed to evaluate the accuracy of spatial relationships in text-to-image generation, achieving performance that even surpasses leading proprietary models on spatial evaluation. We further demonstrate that this reward model effectively enables online reinforcement learning for the complex spatial generation. Extensive experiments across multiple benchmarks show that our specialized reward model yields significant and consistent gains in spatial understanding for image generation. |
| title | Enhancing Spatial Understanding in Image Generation via Reward Modeling |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.24233 |