Administrative Decentralization in Edge-Cloud Multi-Agent for Mobile Automation

2026-04-08

AI Summary

This work addresses the limitations of existing edge-cloud collaborative frameworks, which suffer from delayed edge responsiveness and insufficient autonomous decision-making due to centralized cloud-based control. To overcome these issues, the authors propose AdecPilot, a novel framework that introduces administrative decentralization into edge-cloud multi-agent systems for the first time. By decoupling high-level policy design from low-level tactical execution, AdecPilot endows edge agents with autonomous planning and self-correction capabilities. Key innovations include UI-agnostic cloud-based policy generation, a dual-mode edge execution team, and a hierarchical implicit termination protocol, collectively ensuring task determinism and mitigating hallucinatory behaviors. Experimental results demonstrate that the proposed approach improves task success rates by 21.7%, reduces cloud token consumption by 37.5% compared to EcoAgent, and decreases end-to-end latency by 88.9% relative to CORE.

Abstract

Collaborative edge-cloud frameworks have emerged as the main- stream paradigm for mobile automation, mitigating the latency and privacy risks inherent to monolithic cloud agents. However, existing approaches centralize administration in the cloud while relegating the device to passive execution, inducing a cognitive lag regard- ing real-time UI dynamics. To tackle this, we introduce AdecPilot by applying the principle of administrative decentralization to the edge-cloud multi-agent framework, which redefines edge agency by decoupling high-level strategic designing from tactical grounding. AdecPilot integrates a UI-agnostic cloud designer generating ab- stract milestones with a bimodal edge team capable of autonomous tactical planning and self-correction without cloud intervention. Furthermore, AdecPilot employs a Hierarchical Implicit Termi- nation protocol to enforce deterministic stops and prevent post- completion hallucinations. Extensive experiments demonstrate pro- posed approach improves task success rate by 21.7% while reducing cloud token consumption by 37.5% against EcoAgent and decreas- ing end to end latency by 88.9% against CORE. The source code is available at https://anonymous.4open.science/r/Anonymous_code- B8AB.

Problem

• administrative decentralization
• edge-cloud multi-agent
• mobile automation
• cognitive lag
• real-time UI dynamics

Innovation

• administrative decentralization
• edge-cloud multi-agent
• tactical grounding
• hierarchical implicit termination
• mobile automation