ZonUI-3B
Collection
ZonUI-3B — A lightweight GUI grounding model optimized for high-resolution screens, trained with just 24K examples on a single RTX 4090. • 4 items • Updated • 2
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
This dataset is a curated 8K-sample subset from the original ShowUI-web dataset, as mentioned in our paper. It contributes to the training of GUI grounding models, with a focus on realistic web user interfaces collected from diverse websites.
This dataset is one of the sources used in our paper, ZonUI-3B: A Lightweight Vision-Language Model for Cross-Resolution GUI Grounding. The ShowUI-web subset provides real-world web UI data, helping the model develop grounding capability in diverse and responsive web contexts.
If you use this dataset, please cite the following paper:
@misc{hsieh2025zonui3b,
title = {ZonUI-3B: A Lightweight Vision-Language Model for Cross-Resolution GUI Grounding},
author = {Hsieh, ZongHan and Wei, Tzer-Jen and Yang, ShengJing},
year = {2025},
howpublished = {\url{https://arxiv.org/abs/2506.23491}},
note = {arXiv:2506.23491 [cs.CV], version 2, last revised 1 Jul 2025}
}