Yitao Liu

Dear Authors, Thank you for this brilliant work. I want to do some analysis on the trajectories of different methods in your paper (e.g. FLAN-T5, GPT-4, SeeAct with different grounding...

Unstablility of gitlab website

1

Dear Authors, I deployed the Gitlab website locally, but I found that it is unstable, e.g., if I keep visiting it, sometimes it works normally, while sometimes it reports 502....

Pop-up windows and Combobox can not be viewed

1

Dear authors, Thanks for your nice work and quick follow-ups for issues. I found two problems when using the benchmark, especially on the map website: 1. The pop-up windows, e.g.,...

Problem when redirecting to a new page in a new tab

2

Dear authors, Thanks for your brilliant work. When I tried to use this benchmark, I found two possible problems: 1. if some action, e.g. clicking a link, will open a...

Add ignore error flag

1

During the execution of function 'generate_schema_augmented_prompt_main', code execution errors may appear(it actually appears when I use this dataset, which hinders prompt generation). So I think it is necessary to add...