Imbalance in Balance: Online Concept Balancing in Generation Models
Abstract
In visual generation tasks, the responses and combinations of complex concepts often lack stability and are error-prone, which remains an under-explored area. In this paper, we attempt to explore the causal factors for poor concept responses through elaborately designed experiments. We also design a concept-wise equalization loss function (IMBA loss) to address this issue. Our proposed method is online, eliminating the need for offline dataset processing, and requires minimal code changes. In our newly proposed complex concept benchmark Inert-CompBench and two other public test sets, our method significantly enhances the concept response capability of baseline models and yields highly competitive results with only a few codes.
Cite
Text
Shi et al. "Imbalance in Balance: Online Concept Balancing in Generation Models." International Conference on Computer Vision, 2025.Markdown
[Shi et al. "Imbalance in Balance: Online Concept Balancing in Generation Models." International Conference on Computer Vision, 2025.](https://mlanthology.org/iccv/2025/shi2025iccv-imbalance/)BibTeX
@inproceedings{shi2025iccv-imbalance,
title = {{Imbalance in Balance: Online Concept Balancing in Generation Models}},
author = {Shi, Yukai and Ou, Jiarong and Chen, Rui and Yang, Haotian and Wang, Jiahao and Tao, Xin and Wan, Pengfei and Zhang, Di and Gai, Kun},
booktitle = {International Conference on Computer Vision},
year = {2025},
pages = {17432-17442},
url = {https://mlanthology.org/iccv/2025/shi2025iccv-imbalance/}
}