mlsanityData to 14 Oct 2026

Spine-Branch Coordination for Multi-agent Computer Use

arXiv 2608.22077 ↗First posted 22 Aug 2026cs.CL cs.MAcited by 0

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Computer use agents (CUAs) are increasingly deployed as multi-agent systems that decompose a task into multiple subtasks executed across parallel virtual machines (VMs). However, a critical physical bottleneck is that the state of two VMs cannot be merged. Previous systems handle this ad-hoc rather than treating it as a first-class concern. We propose Spine-Branch Coordination for multi-agent computer use, a framework that decomposes a task into a "spine-branch" graph, where the spine carries the main task flow with continuous VM state and branch tasks execute in parallel to collect information the spine needs to complete the task. Branch VMs are discarded once their tasks finish, so no VM merging ever occurs. Experiments show that on 200 long-horizon tasks from Odysseys and across three CUA backbones, Spine-Branch improves success rate over the baseline system by 6.0% to 16.5%, while reducing per-task cost by 34% to 70%, indicating that explicitly modeling VM-state merging constraint enables multi-agent computer use to scale efficiently.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 31 of its 34 references that are in the map. First posted on arXiv 22 Aug 2026.

Builds on

31 of 34 references.

Closest ancestors · 6

Same direction, newest first

  1. 2026Decentralized Multi-Agent Systems with Shared Context9 Jun
  2. 2026AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration3 Feb
  3. 2025Multi-Agent Coordination across Diverse Applications: A Survey20 Feb
  4. 2024WebPilot: A Versatile and Autonomous Multi-Agent System for Web Task Execution with Strategic Exploration28 Aug
  5. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework1 Aug
  6. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 25

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  2. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  4. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  5. 2025Qwen3-VL Technical Reportcited by 2,793
  6. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 2,006
  7. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,988
  8. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  9. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agentscited by 1,251
  10. 2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,195
  11. 2023ChatDev: Communicative Agents for Software Developmentcited by 919
  12. 2024GPT-4V(ision) is a Generalist Web Agent, if Groundedcited by 563
13 more
  1. 2024AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agentscited by 468
  2. 2024WebVoyager: Building an End-to-End Web Agent with Large Multimodal Modelscited by 438
  3. 2018Reinforcement Learning on Web Interfaces Using Workflow-Guided Explorationcited by 284
  4. 2024Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scalecited by 223
  5. 2024Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMscited by 176
  6. 2025An Illusion of Progress? Assessing the Current State of Web Agentscited by 174
  7. 2025Agent S2: A Compositional Generalist-Specialist Framework for Computer Use Agentscited by 159
  8. 2026OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Taskscited by 40
  9. 2025CoAct-1: Computer-using Multi-Agent System with Coding Actionscited by 35
  10. 2025Fara-7B: An Efficient Agentic Model for Computer Usecited by 35
  11. 2026Odysseys: Benchmarking Web Agents on Realistic Long Horizon Taskscited by 21
  12. 2026iOSWorld: A Benchmark for Personally Intelligent Phone Agentscited by 7
  13. 2026Multi-Agent Computer Usecited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use Agents
  2. 2026DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving
  3. 2026When Parallelism Pays Off: Cohesion-Aware Task Partitioning for Multi-Agent Coding
  4. 2026AdaptOrch: Task-Adaptive Multi-Agent Orchestration in the Era of LLM Performance Convergence
  5. 2026Attention Tax, Handoff Tax: A Stylised Model of When Multi-Agent LLM Systems Help
  6. 2026Small Model as Master Orchestrator: Learning Unified Agent-Tool Orchestration with Parallel Subtask Decomposition
  7. 2026Architectural Design, Not Only Model Intelligence, Governs Multi-Agent LLM Performance
  8. 2025DynTaskMAS: A Dynamic Task Graph-driven Framework for Asynchronous and Parallel LLM-based Multi-Agent Systems