CompoWorld: Compositional Environment Scaling for General Agents
Authors:Xiao-Wen Yang, Weiyi Xu, Wen Da, Hang Xu, Canwei Li, Hong-Jie You, Pusen Dong, Yucheng Zeng, Zhaokai Luo, Yu-Feng Li, Yao Hu, Mu Chuan
Abstract:Automatically generated environments provide a scalable source of interaction data for training general agents. However, existing approaches mainly generate tasks within a single environment, while real-world workflows require agents to connect information and actions across multiple services. We introduce Compositional Environment Scaling (\textbf{CompoWorld}), which expands the task space by composing a finite library of reusable services. Coding agents turn tool specifications into verified services with typed states and shared interfaces, while a world model handles tools that cannot be reliably implemented. A random-walk procedure connects services through dependency graphs, enabling the generation and verification of tasks that require information to flow across services. Verified trajectories support supervised fine-tuning (SFT), while our Completion-Focused Rubric Reward guides reinforcement learning (RL) toward full task completion by emphasizing criteria with lower pass rates within each rollout group. We construct 448 services exposing 10,130 tools and use 3K SFT trajectories and 1K RL tasks to train Qwen3.6-35B-A3B. Experimental results show that CompoWorld improves on its backbone by 9.17 points on average across eight benchmarks. On AutomationBench, it surpasses frontier models such as Claude Opus 4.6 and leads all compared agent-specialized 35B-A3B models.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.33665 [cs.AI] |
| (or arXiv:2609.33665v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.33665 arXiv-issued DOI via DataCite |
Submission history
From: Xiao-Wen Yang [view email]
[v1]
Sun, 27 Sep 2026 15:25:01 UTC (835 KB)
[v2]
Wed, 30 Sep 2026 08:08:26 UTC (833 KB)