RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents

Hugging Face Daily Papers Papers
computer-use-agents scalable-framework verification ai-benchmark agent-environments

Summary

RecreationWorld introduces a scalable and verifiable framework for hybrid computer-use agents, providing environments across five platforms and a benchmark for evaluation.

Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development through code and the command line. Real digital work requires both, interleaved rather than stacked end to end. We study hybrid CUAs that autonomously decide when to explore an interface, implement software, and run and visually verify their artifacts. We introduce RecreationWorld, a five-platform framework built around recreation: given a running reference, an agent must discover its behavior and build a faithful implementation with no prescribed workflow. RecreationWorld provides reproducible environments on Ubuntu, macOS, Windows, Android, and Web, plus a unified harness with native GUI control and coding tools. The running reference serves as an oracle for hidden behavioral tests, providing execution-grounded rewards. We scale trajectory generation with high-quality open-source applications. Models trained on these trajectories improve across five out-of-distribution coding and hybrid computer-use benchmarks and more frequently verify their rendered outputs, providing evidence of transfer beyond recreation. For held-out evaluation, we introduce RecreationBench, comprising 250 diverse tasks across domains and platforms. Reference-grounded programmatic and visual assertions cover action-conditioned outcomes at multiple interaction depths; each is validated on the reference and by human reviewers before the suite is frozen for automatic scoring. GPT-6 Astra leads at 58.1% overall, but passes all programmatic tests on just 2.8% of tasks. Agents reproduce static interface structure more reliably than interactions and computed outputs, while generated applications remain smaller and more monolithic than their references. We release the benchmark, environments, and test suites.
Original Article
View Cached Full Text

Cached at: 09/21/26, 03:19 AM

Paper page - RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents

Source: https://huggingface.co/papers/2609.22000 Published on Sep 18

#1 Paper of the day Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

Computer-useagents(CUAs)haveadvancedalongtwoseparatelines:graphicalinteractionandsoftwaredevelopmentthroughcodeandthecommandline.Realdigitalworkrequiresboth,interleavedratherthanstackedendtoend.WestudyhybridCUAsthatautonomouslydecidewhentoexploreaninterface,implementsoftware,andrunandvisuallyverifytheirartifacts.WeintroduceRecreationWorld,afive-platformframeworkbuiltaroundrecreation:givenarunningreference,anagentmustdiscoveritsbehaviorandbuildafaithfulimplementationwithnoprescribedworkflow.RecreationWorldprovidesreproducibleenvironmentsonUbuntu,macOS,Windows,Android,andWeb,plusaunifiedharnesswithnativeGUIcontrolandcodingtools.Therunningreferenceservesasanoracleforhiddenbehavioraltests,providingexecution-groundedrewards.Wescaletrajectorygenerationwithhigh-qualityopen-sourceapplications.Modelstrainedonthesetrajectoriesimproveacrossfiveout-of-distributioncodingandhybridcomputer-usebenchmarksandmorefrequentlyverifytheirrenderedoutputs,providingevidenceoftransferbeyondrecreation.Forheld-outevaluation,weintroduceRecreationBench,comprising250diversetasksacrossdomainsandplatforms.Reference-groundedprogrammaticandvisualassertionscoveraction-conditionedoutcomesatmultipleinteractiondepths;eachisvalidatedonthereferenceandbyhumanreviewersbeforethesuiteisfrozenforautomaticscoring.GPT-6Astraleadsat58.1%overall,butpassesallprogrammatictestsonjust2.8%oftasks.Agentsreproducestaticinterfacestructuremorereliablythaninteractionsandcomputedoutputs,whilegeneratedapplicationsremainsmallerandmoremonolithicthantheirreferences.Wereleasethebenchmark,environments,andtestsuites.

View arXiv pageView PDFGitHub1Add to collection

Get this paper in your agent:

hf papers read 2609\.22000

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.22000 in a model README.md to link it from this page.

Datasets citing this paper1

#### Qwen/RecreationBench Viewer• Updated35 minutes ago • 500 • 1

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.22000 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

OpenComputer: Verifiable Software Worlds for Computer-Use Agents

Hugging Face Daily Papers

OpenComputer presents a framework for creating verifiable software environments for computer-use agents, integrating state verifiers, self-improving verification layers, task synthesis, and evaluation systems across 33 desktop applications. Experiments show its verifiers align better with human judgment than LLM-as-judge, and frontier agents struggle with end-to-end completion.