HaoranLiu/DPO-Qwen3-MobileGym download history

HaoranLiu/DPO-Qwen3-MobileGym is an image text to text dataset on the Hugging Face Hub. In the last 30 days it was downloaded 22 times (4 in the last 7 days), and 123 times in total. It ranks #349,066 among datasets by monthly downloads.

DPO-Qwen3-MobileGym Trajectory-level DPO preference pairs for mobile GUI agents, rendered and tokenized for Qwen/Qwen3-VL-4B-Instruct on the mobilegym environment. What this is 336 (chosen, rejected) trajectory pairs over 89 mobilegym tasks. Both sides of a pair share a

Models trained on DPO-Qwen3-MobileGym

1 models list it as training data.

Open HaoranLiu/DPO-Qwen3-MobileGym on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.