More details about OSWorld running script.

Open jackroos opened this issue 10 months ago • 3 comments

Could you provide the argument values in run_uitars.py for reproducing OSWorld results?

Feb 18 '25 09:02 jackroos

Hi, the argument values are here: https://github.com/xlang-ai/OSWorld/blob/main/mm_agents/uitars_agent.py#L404

Feb 20 '25 10:02 pooruss

Hi, the argument values are here: https://github.com/xlang-ai/OSWorld/blob/main/mm_agents/uitars_agent.py#L404

https://github.com/xlang-ai/OSWorld/blob/884676cebc4300983cf1285177b30dcc8ef25746/mm_agents/uitars_agent.py#L413

The observation is screenshot + a11y tree? It is confusing since UI-TARS is trained on screenshot-only observation.

Feb 21 '25 07:02 jackroos

Any progress on this?

Mar 12 '25 15:03 szydev6