Aug 2026· Tsinghua Science and Technology· 0 citations
TL;DR
This work introduces a diffusion model as a generative prior to produce high-quality global deployment plans, effectively avoiding the local-optima problem common in conventional reinforcement learning.
Abstract
Deploying Large Language Model (LLM)-based AI agents at the network edge enables real-time task handling in 6G networks. However, resource heterogeneity and user mobility necessitate agent migration to maintain Quality of Service (QoS). Unlike stateless containers, AI agents encapsulate invocation histories, planning contexts, and memory stores, making their migration considerably more complex. To address this issue, we propose AMD, an adaptive AI Agent Migration framework that integrates conditional Diffusion models with reinforcement learning to jointly optimize latency and resource utilization under dynamic edge environments. Specifically, we introduce a diffusion model as a generative prior to produce high-quality global deployment plans, effectively avoiding the local-optima problem common in conventional reinforcement learning. A reinforcement learning-based module then performs online policy refinement to accommodate real-time environmental variations. We implement AMD on a distributed system built upon AgentScope and validate it across geographically distributed edge servers. Experimental results demonstrate that AMD reduces the average task latency by 3.9%–27.6% and improves resource utilization by up to 53.3% compared to baseline strategies.
A Graph-Structured Experiential Memory framework for multi-agent coordination in dynamic manufacturing that encodes historical coordination episodes as heterogeneous relational graphs that capture task dependencies, machine states, and inter-agent collaboration patterns is proposed.
Chengxiao Dai, Zhanhui Lin, Zhaokun Yan et al.· 0 citations
MANTA, a framework for Multi-Agent Network Topology Adaptation that enables communication structures to self-evolve at inference time, is introduced and shows that inference-time self-improvement can extend to the architecture of collaboration itself.
M. Huang, Jerry Wang, Yi-Cheng Lai et al.· 0 citations
This research bridges theoretical foundations of reinforcement learning and graph-based memory with autonomous agent workflows, and offers a practical, scalable reference framework for developing artificial intelligence technologies in complex, multi-step autonomous systems.
High-velocity workloads and intricate task dependencies inherent in distributed stream-processing systems pose a fundamental challenge to efficient resource allocation. Traditional heuristic and single-agent reinforcement learning (RL) schedulers frequently fail to recognize these complex network and data-flow interactions, leading to severe resource fragmentation and catastrophic tail latency spikes. In order to accomplish coordinated, low-latency scheduling, we propose a Topology-Aware Multi-Agent Reinforcement Learning (TAMARL) framework utilizing a Centralized Training and Decentralized Execution (CTDE) architecture. TAMARL allows distributed agents to optimize task placement across heterogeneous cluster nodes and prevent backpressure cascades by integrating topology-aware state representations. We evaluate TAMARL on a production-grade cloud-native stack leveraging Apache Flink and Kubernetes. Compared to state-of-the-art baselines across six demanding stress-test scenarios, experimental evaluations demonstrate that TAMARL improves Service Level Objective (SLO) attainment by 27% while reducing P99 tail latency by up to 68%. Additionally, TAMARL maintains stable, resilient performance under 90% cluster utilization while securing 95% network locality.
Sunday J. Awine, Jinwei Liu· IEEE International Conferenc...· 0 citations
Real-time data streaming systems increasingly operate under highly variable workloads, heterogeneous data sources, latency constraints, and frequent service disruptions. Conventional stream-processing architectures generally depend on predefined routing, static resource allocation, and centralized coordination, which can limit their ability to adapt when event rates, computational requirements, or infrastructure conditions change rapidly. This paper proposes an Adaptive Multi-Agent AI Framework for Real-Time Data Streaming with Enhanced Scalability and Resilience, in which autonomous AI agents collaboratively perform stream monitoring, workload classification, task allocation, resource adaptation, anomaly detection, and recovery. The theoretical foundation combines multi-agent coordination with contextual representation, long-document processing, memory management, and adaptive decision-making. Prior work on aspect-controllable summarization demonstrates the value of controlling computational objectives according to task requirements, while studies of coreference, lexical chains, and entity-based coherence emphasize the importance of preserving relationships across distributed information units (Amplayo, Angelidis, & Lapata, 2021; Baldwin & Morton, 1998; Barzilay & Elhadad, 1997; Barzilay & Lapata, 2005). Long-context language modeling further motivates mechanisms capable of retaining relevant information over extended streaming windows (Beltagy, Peters, & Cohan, 2020). The proposed framework extends these principles to adaptive streaming environments and aligns with recent multi-agent event-streaming research emphasizing resiliency and scalability (Reddy et al., 2026). Analytical findings indicate that decentralized agent specialization, shared contextual state, adaptive workload redistribution, and failure-aware coordination can provide a stronger basis for resilient streaming than static pipelines. The paper also identifies trade-offs involving coordination overhead, state consistency, model complexity, and resource consumption.
Nethmi Perera, Kasun Fernando· American Journal Of Applied...· 0 citations
This paper proposes a multi-agent reinforcement learning (MARL) framework for TSN scheduling, where each TSN queue is modeled as an autonomous agent and the Heterogeneous-Agent Proximal Policy Optimization (HAPPO) algorithm is employed to explicitly model inter-agent dependencies and jointly optimize service delivery across queues.
Marcos Carvalho, Fatih Temiz, Shavbo Salehi et al.· 0 citations