Aniket Rege, Zinnia Nie, Mahesh Ramesh, Unmesh Raskar, Zhuoran Yu, Aditya Kusupati, Yong Jae Lee, Ramya Korlakai Vinayak
Instructions to prepare the benchmark dataset are provided in prepare_data/
T2I inference, metric computation on generated images, and plotting code is provided in experiments/ - Coming Soon
Details on designing and launching the user study with Qualtrics and Prolific respectively are provided in user_study/ - Coming Soon
If you find our work useful in your research, please consider citing:
@misc{rege2025cure,
title={CuRe: Cultural Gaps in the Long Tail of Text-to-Image Systems},
author={Aniket Rege and Zinnia Nie and Mahesh Ramesh and Unmesh Raskar and Zhuoran Yu and Aditya Kusupati and Yong Jae Lee and Ramya Korlakai Vinayak},
year={2025},
eprint={2506.08071},
url={https://arxiv.org/abs/2506.08071},
}