Much of the previous work towards digital agents for graphical user interfaces (GUIs) has relied on text-based representations (derived from HTML or other structured data sources), which are not always readily available. These input representations have been often coupled with custom, task-specific action spaces. This paper focuses on creating agents that interact with the digital world using the same conceptual interface that humans commonly use -- via pixel-based screenshots and a generic action space corresponding to keyboard and mouse actions. Building upon recent progress in pixel-based pretraining, we show, for the first time, that it is possible for such agents to outperform human crowdworkers on the MiniWob++ benchmark of GUI-based instruction following tasks.
@article{arxiv.2306.00245,
title = {From Pixels to UI Actions: Learning to Follow Instructions via Graphical User Interfaces},
author = {Peter Shaw and Mandar Joshi and James Cohan and Jonathan Berant and Panupong Pasupat and Hexiang Hu and Urvashi Khandelwal and Kenton Lee and Kristina Toutanova},
journal= {arXiv preprint arXiv:2306.00245},
year = {2023}
}