# StateAct: Program State, before Pixels, for Long-Horizon Computer-Use Agents

arXiv: https://arxiv.org/abs/2607.22798 · HTML: https://mlsanity.com/paper/2607.22798/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 21 references that are in the map.
- First posted on arXiv 24 Jul 2026.

## Builds on (most cited first, 19 of 19)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [UI-TARS: Pioneering Automated GUI Interaction with Native Agents](https://mlsanity.com/paper/2501.12326/) (2025)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents](https://mlsanity.com/paper/2405.14573/) (2024)
- [Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction](https://mlsanity.com/paper/2412.04454/) (2024)
- [Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale](https://mlsanity.com/paper/2409.08264/) (2024)
- [GTA1: GUI Test-time Scaling Agent](https://mlsanity.com/paper/2507.05791/) (2025)
- [MobileWorld: Benchmarking Autonomous Mobile Agents in Agent-User Interactive and MCP-Augmented Environments](https://mlsanity.com/paper/2512.19432/) (2025)
- [ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents](https://mlsanity.com/paper/2508.14040/) (2025)
- [UFO2: The Desktop AgentOS](https://mlsanity.com/paper/2504.14603/) (2025)
- [OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks](https://mlsanity.com/paper/2606.29537/) (2026)
- [MCPWorld: A Unified Benchmarking Testbed for API, GUI, and Hybrid Computer Use Agents](https://mlsanity.com/paper/2506.07672/) (2025)
- [CoAct-1: Computer-using Multi-Agent System with Coding Actions](https://mlsanity.com/paper/2508.03923/) (2025)
- [UltraCUA: A Foundation Model for Computer Use Agents with Hybrid Action](https://mlsanity.com/paper/2510.17790/) (2025)
- [Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification](https://mlsanity.com/paper/2507.11662/) (2025)
- [OpenComputer: Verifiable Software Worlds for Computer-Use Agents](https://mlsanity.com/paper/2605.19769/) (2026)

## Built on by (most cited first, 3 of 3)

- [RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents](https://mlsanity.com/paper/2609.22000/) (2026)
- [GUI-HARVEST: Self-Improving GUI Agents through Evidence-Driven Harness Evolution](https://mlsanity.com/paper/2610.00948/) (2026)
- [CUAWright: A Minimal Unified Interface for Digital Agents](https://mlsanity.com/paper/2610.04116/) (2026)

## Abstract (arXiv)

Computer-use agents are usually improved by strengthening perception: better models for reading a screenshot and choosing where to click. Yet a screenshot is only a lossy rendering of the underlying program state, e.g., the files, application backends, and DOM that hold the task data. Different states can produce the same pixels, while code can inspect and modify that state directly. StateAct is a code-first, multi-agent harness built around this distinction. Its main agent works directly with program state by using code, while a dedicated GUI subagent handles screenshot-and-click interaction on the few subgoals that need it, just 28 of 108 tasks and 1.1% of main-agent steps. The same direct access to program state also supports verification: an independent finish gate double-checks the saved result for structural failures, e.g., output that is missing, unsaved, or written to the wrong path. To stay on track over hundreds of steps, the main agent hands subgoals to fresh subagents, keeping its own context focused. On OSWorld 2.0, StateAct lifts Claude Opus 4.8 from 20.6% to 26.9% on binary success, and from 54.8% to 61.6% on partial success, at ~ 9x lower cost per task than the same model driven by screenshots alone; a code-only variant with no GUI subagent reaches only 45.9% partial, below that screenshot-based baseline's 54.8%. In general, grounding action, verification, and memory in state, what we call state-grounding, shifts the main bottleneck from perception toward reasoning: failures depend more on what the agent thinks than on what it sees.