AgentVNE: LLM-Augmented Graph Reinforcement Learning for Affinity-Aware Multi-Agent Placement in Edge Agentic AI

2026-01-05

arXiv.org

AI Summary

This work addresses the challenge of efficiently deploying multi-agent services with complex topological dependencies and affinity constraints in resource-constrained edge environments, where traditional virtual network embedding (VNE) methods struggle. The authors propose AgentVNE, a novel framework that, for the first time, integrates large language models (LLMs) to parse multi-agent workflows and extract implicit semantic constraints. These constraints are jointly reasoned with topological structures through a resource similarity-aware graph neural network. The framework further employs a hybrid strategy combining pretraining and proximal policy optimization (PPO)-based reinforcement learning to dynamically optimize agent placement. Experimental results demonstrate that under high-load scenarios, AgentVNE reduces communication latency to less than 40% of baseline methods while improving service acceptance rates by 5%–10%.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Adversarial Agents
• Machine Learning: Hardware-aware ML

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

The Internet of Agents is propelling edge computing toward agentic AI and edge general intelligence (EGI). However, deploying multi-agent service (MAS) on resource-constrained edge infrastructure presents severe challenges. MAS service workflows are driven by complex cross-node interactions, dynamic memory accumulation, and collaborative tool usage. Exhibiting chain-like topological dependencies and strict affinity constraints, these workflows demand real-time responsiveness that exceeds the capabilities of traditional VNE algorithms designed for static resources. To address this, we propose AgentVNE, a cloud-edge collaborative framework utilizing a dual-layer architecture. First, AgentVNE employs a large language model (LLM) to identify implicit semantic constraints and generate affinity-based resource augmentation to resolve physical dependency issues. Second, it constructs a resource similarity-aware neural network, utilizing a pre-training and PPO fine-tuning strategy to precisely capture topological similarities between dynamic workflows and heterogeneous networks. By coupling semantic perception with topological reasoning, this mechanism effectively bridges the gap between dynamic service requirements and physical infrastructure. Simulation results demonstrate that AgentVNE reduces workflow communication latency to less than 40% of baselines and improves the service acceptance rate by approximately 5%-10% under high-load scenarios. Ultimately, this work provides a foundational solution for the semantic-aware deployment of agentic AI.

Problem

• multi-agent placement
• edge computing
• affinity constraints
• virtual network embedding
• agentic AI

Innovation

• LLM-Augmented
• Graph Reinforcement Learning
• Affinity-Aware Placement
• Edge Agentic AI
• Resource Similarity