跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-19AI 評分48

WildAWM:合成電腦使用智能體環境與任務

Agent World Model in the Wild: Computer-Use Agent Environment Synthesis

AI 導讀

WildAWM 透過合成可控的電腦使用環境與任務,訓練通用智能體,並提升其在分佈外基準上的表現。該流程支援 GUI、MCP 服務及 CLI,合成了 341 組長期環境與任務配對;部分任務需要超過 100 個智能體步驟,GPT-6-Astra 僅能完成其中 60%。研究團隊只用合成環境訓練兩個不同規模的 Qwen 模型,並在涵蓋網頁、CLI 及電腦使用智能體的 8 個分佈外基準上取得顯著提升。

正文

Published on Oct 19

#3 Paper of the day

Authors:

,

,

,

,

,

,

,

,

,

,

Abstract

WildAWM synthesizes controllable computer-use environments and tasks to train general agents via interaction, improving out-of-distribution performance across web, CLI, and GUI benchmarks.

Agent learning is shifting from learning from human data toward learning through interaction with the world. However, scaling such learning directly in real-world computer systems is challenging and can introduce substantial security risks. We therefore propose Agent World Model in the Wild (WildAWM), a scalable pipeline for synthesizing controllable environments and tasks for training general computer-use agents (CUAs). Whereas existing approaches typically specialize in a single agent interaction interface, we construct executable environments that jointly support graphical user interfaces, model context protocol services, and command-line interfaces, allowing agents to flexibly combine visual interaction, tool use, and code execution. WildAWM is entirely powered by open-source models, demonstrating that high-quality computer-use environments can be generated without proprietary frontier models. Beyond environments, we address three recurring challenges in task synthesis: ensuring solvability, maintaining sufficient difficulty, and producing reliable verifiers, by introducing execution-based quality gates that combine heterogeneous agent rollouts, cross-judge consistency checks, and counterfactual probes of judgment robustness. The resulting pipeline synthesizes 341 long-horizon environment-task pairs that can require over 100 agent steps to solve and remain challenging even for frontier models, with GPT-6-Astra solving only 60% of them. We train two Qwen models of different sizes exclusively in these synthetic worlds. Both models achieve substantial gains across 8 out-of-distribution benchmarks spanning web, claw, and computer-use agents, demonstrating that learning in WildAWM's synthesized worlds generalizes broadly to unseen computer-use tasks. The code is publicly available at https://github.com/MSR-Orchard/WildAWM.

View arXiv page View PDF Project page GitHub 50 Add to collection

Get this paper in your agent:

hf papers read 2610.26828

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 2

ChilleD/WildAWM-35B-A3B

36B •

Updated 2 days ago

•

40

ChilleD/WildAWM-4B

5B •

Updated 2 days ago

•

35

Datasets citing this paper 1

ChilleD/WildAWM

Preview •

Updated 1 day ago

•

213

•

1

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.26828 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