Mage-VL-Cowork:從演進後的 Harness 學習混合式電腦操作的完整訓練方案
Mage-VL-Cowork: A Full Training Recipe to Learn Hybrid Computer Use From Evolved Harnesses
Mage-VL-Cowork 是一款 4B 混合式電腦操作智能體,經五階段訓練後在 OSWorld-Verified 取得 66.0%,達到同參數規模的最先進表現。Mage-CUA-Forge 合成了 293,806 項桌面及網頁任務,涵蓋 6,915 項技能,並將重複任務開支減少 82.6%。
Published on Oct 29
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
Mage-VL-Cowork is a 4B hybrid computer-use agent trained via a progressive curriculum and scalable synthetic data generation, achieving state-of-the-art performance on OSWorld-Verified with efficient reasoning.
Effective computer-use agents (CUAs) should seamlessly combine visual GUI interaction with code and terminal execution within a single policy. We present Mage-VL-Cowork, a 4B hybrid CUA trained from a general multimodally aligned checkpoint through an end-to-end, progressive five-stage curriculum across midtraining, SFT, and RL. To power this curriculum at scale, Mage-CUA-Forge replaces per-task agent construction with reusable, skill-conditioned generation workflows and application-specific preparers, synthesizing 293,806 desktop and web tasks across 6,915 skills while reducing recurring per-task expenditure by 82.6%. Rather than directly distilling a default teacher, we evolve the teacher harness and instantiate specialized CUA harnesses to yield high-quality trajectories alongside stage-specific supervision. Mage-VL-Cowork-4B finally climbs to 66.0% on OSWorld-Verified through different training stages with more efficient reasoning cost, achieving state-of-the-art performance at its parameter size and pushing forward the frontier of on-device intelligence.
View arXiv page View PDF Project page Add to collection
Get this paper in your agent:
hf papers read 2610.40747
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.40747 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.40747 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.40747 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co