Current multi-subject customization approaches encounter two critical challenges: the difficulty in acquiring diverse multi-subject training data, and attribute entanglement across different subjects. To bridge these gaps, we propose MUSAR - a simple yet effective framework to achieve robust multi-subject customization while requiring only single-subject training data. Firstly, to break the data limitation, we introduce debiased diptych learning. It constructs diptych training pairs from single-subject images to facilitate multi-subject learning, while actively correcting the distribution bias introduced by diptych construction via static attention routing and dual-branch LoRA. Secondly, to eliminate cross-subject entanglement, we introduce dynamic attention routing mechanism, which adaptively establishes bijective mappings between generated images and conditional subjects. This design not only achieves decoupling of multi-subject representations but also maintains scalable generalization performance with increasing reference subjects. Comprehensive experiments demonstrate that our MUSAR outperforms existing methods - even those trained on multi-subject dataset - in image quality, subject consistency, and interaction naturalness, despite requiring only single-subject dataset.
@article{arxiv.2505.02823,
title = {MUSAR: Exploring Multi-Subject Customization from Single-Subject Dataset via Attention Routing},
author = {Zinan Guo and Pengze Zhang and Yanze Wu and Chong Mou and Songtao Zhao and Qian He},
journal= {arXiv preprint arXiv:2505.02823},
year = {2025}
}
Comments
Project page at https://github.com/guozinan126/MUSAR