The model stayed 27B. The score didn't.

Mine workflows from tutorials into verified tasks; retry the rollout from a checkpoint. 80.2 on OSWorld Verified. 1 2 3

1
CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements
arxiv.org
2
The paper's full text, with every table behind these numbers
arxiv.org
3
Its page on the Hugging Face daily papers board
huggingface.co

This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.

Comments

1/4

2026-11-03

Mine workflows from tutorials into verified tasks; retry the rollout from a checkpoint. 80.2 on OSWorld Verified. 1 2 3

1
CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements
arxiv.org
2
The paper's full text, with every table behind these numbers
arxiv.org
3
Its page on the Hugging Face daily papers board
huggingface.co

This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.