Oriel — AI & technology worth noticing, every day.

CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements

Paper · https://arxiv.org/abs/2611.03126

Despite recent advances in CUAs, their development remains constrained by the limited diversity, controllability, and reliability of interaction data. Moreover, conventional optimization pipelines lack iterative feedback, while existing recursive improvement approaches often rely on self-generated experiences, limiting ability expansion and yielding diminishing returns. To address these challenges, we introduce CUA-Master, a nested double-loop recursive improvement framework. At the system level, an outer world-data-model flywheel continuously incorporates open-world corpus and leverages evolving model weaknesses to guide subsequent data construction and optimization iterations. At the data level, an inner flywheel couples open-world seeding and autocatalytic propagation. The former mines authentic workflows as seed tasks from instructional videos, and employs visual-semantic dual-path trajectory reconstruction; The latter transforms these seeds into reusable primitives for controllable synthesis of diverse, complex, and verifiable tasks. To enable efficient and reliable trajectory collection, we further introduce a feedback-guided snapshot-rollback retry mechanism in various real environments. Extensive experiments demonstrate that CUA-Master-27B achieves competitive performance across GUI-only and GUI-CLI hybrid benchmarks, scoring 80.2 on OSWorld Verified, 36.9 on OSWorld 2.0, 43.9 on WeaveBench, and 25.0 on Agents' Last Exam. Our framework also substantially improves the quality and efficiency of data synthesis, and exhibits continuous gains across recursive iterations.

First seen by Oriel on .

AI summary

CUA-Master introduces a nested recursive-improvement framework for computer-use agents, combining open-world data collection and synthesis to improve task diversity and agent performance.

  • Its outer loop uses open-world corpora and model weaknesses to guide successive data construction and optimization.
  • Its inner loop mines workflow seeds from instructional videos, reconstructs trajectories through visual-semantic paths, and propagates reusable primitives into diverse, verifiable tasks.
  • A feedback-guided snapshot-rollback retry mechanism supports trajectory collection across real environments.
  • CUA-Master-27B scores 80.2 on OSWorld Verified, 36.9 on OSWorld 2.0, 43.9 on WeaveBench, and 25.0 on Agents' Last Exam, with gains across recursive iterations.

#computer-use agents #gui automation #data synthesis #recursive learning

Generated by gpt-5.6-sol. Check the original for details.

Metric trends

Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained

Points

Score reported by the source.

  • Hugging Face Papers: 49, latest observation ; -41 over observed period since

Comments

Comment count reported by the source.

  • Hugging Face Papers: 2, latest observation ; +1 over observed period since

Aliases

SourceExternal IDURL
Hugging Face Papers2611.03126https://arxiv.org/abs/2611.03126 Read discussion