UFO$^3$: Weaving the Digital Agent Galaxy

2025-11-14

AI Summary

Current LLM-based agents are predominantly confined to single devices, with cross-device collaboration relying heavily on manual intervention and exhibiting poor robustness. To address this, we propose TaskConstellation—a novel model featuring dynamic Directed Acyclic Graph (DAG) evolution—enabling the first cross-device agent coordination system that supports collective intelligence integration. Our approach models task workflows as distributed DAGs, coordinated by a Constellation Orchestrator and governed by an Agent Interaction Protocol (AIP) for standardized communication. This facilitates automatic task decomposition, asynchronous execution, adaptive fault recovery, and parallel optimization. Evaluated on the NebulaBench benchmark, TaskConstellation achieves an 83.3% subtask completion rate, a 70.9% end-to-end success rate, a parallelism factor of 1.72, and reduces end-to-end latency by 31% compared to serial baselines, while demonstrating resilient fault tolerance and recovery capabilities.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Large language model (LLM)-powered agents are transforming digital devices from passive tools into proactive intelligent collaborators. However, most existing frameworks remain confined to a single OS or device, making cross-device workflows brittle and largely manual. We present UFO$^3$, a system that unifies heterogeneous endpoints, desktops, servers, mobile devices, and edge, into a single orchestration fabric. UFO$^3$ models each user request as a mutable TaskConstellation: a distributed DAG of atomic subtasks (TaskStars) with explicit control and data dependencies (TaskStarLines). The TaskConstellation continuously evolves as results stream in from distributed devices, enabling asynchronous execution, adaptive recovery, and dynamic optimization. A Constellation Orchestrator} executes tasks safely and asynchronously while applying dynamic DAG updates, and the Agent Interaction Protocol (AIP) provides persistent, low-latency channels for reliable task dispatch and result streaming. These designs dissolve the traditional boundaries between devices and platforms, allowing agents to collaborate seamlessly and amplify their collective intelligence. We evaluate UFO$^3$ on NebulaBench, a benchmark of 55 cross-device tasks across 5 machines and 10 categories. UFO$^3$ achieves 83.3% subtask completion, 70.9% task success, exposes parallelism with an average width of 1.72, and reduces end-to-end latency by 31% relative to a sequential baseline. Fault-injection experiments demonstrate graceful degradation and recovery under transient and permanent agent failures. These results show that UFO$^3$ achieves accurate, efficient, and resilient task orchestration across heterogeneous devices, uniting isolated agents into a coherent, adaptive computing fabric that extends across the landscape of ubiquitous computing.

Problem

• Existing agent frameworks are confined to single OS or devices
• Cross-device workflows remain brittle and largely manual
• Traditional boundaries prevent seamless agent collaboration across platforms

Innovation

• Unifies heterogeneous devices into single orchestration fabric
• Models user requests as mutable distributed task constellations
• Provides persistent low-latency channels for reliable task execution