Paper deep dive
Conversational Orchestration for Organic 6G
Masoud Shokrnezhad, Tarik Taleb
Intelligence
Status: succeeded | Model: Gemma-4-26B-A4B | Prompt: intel-v1 | Confidence: 92%
Last extracted: 8/16/2026, 3:11:08 AM
Summary
This paper proposes a lightweight, decentralized conversational orchestration framework for Organic 6G networks, utilizing Large Language Model (LLM)-driven domain agents to manage service provisioning across independently administered domains. The approach addresses the challenges of scalability, simplicity, and agility in dynamic network environments by replacing heavy integration fabrics with an Agent-to-Agent (A2A) overlay. Agents autonomously observe local states, reason via compact models, and exchange summarized reachability advertisements to enable fast placement, scaling, and migration of services while maintaining low control-plane overhead.
Entities (7)
Relation Signals (6)
LLM-driven domain agents â implements â Service Provisioning
confidence 95% ¡ We propose a lightweight, decentralized conversational orchestration framework based on Large Language Model (LLM)-driven domain agents... service provisioning that is simple to operate
LLM-driven domain agents â communicatesvia â Agent-to-Agent (A2A) overlay
confidence 94% ¡ exchanges summaries with neighboring agents over an Agent-to-Agent (A2A) overlay
LLM-driven domain agents â runson â Organic 6G
confidence 93% ¡ The Organic 6G vision... requires... service provisioning... We propose a... framework based on... domain agents.
Compact reasoning model â powers â LLM-driven domain agents
confidence 90% ¡ To meet real-time constraints, we deploy a compact reasoning model... in the proposed agent-based architecture
Organic 6G â includes â Non-Terrestrial Networks (NTN)
confidence 88% ¡ spanning an edge-cloud continuum complemented by non-terrestrial resources... including Non-Terrestrial Networks (NTN) assets
Organic 6G â spans â Edge-cloud continuum
confidence 85% ¡ The Organic 6G vision of a network of networks spanning an edge-cloud continuum
Cypher Suggestions (0)
No Cypher suggestions yet.
Abstract
Abstract:The Organic 6G vision of a network of networks spanning an edge-cloud continuum complemented by non-terrestrial resources requires, to realize its promise, service provisioning that is simple to operate, scalable across independently administered domains, and agile under domain churn (i.e., domains dynamically joining and leaving). Despite advances in cross-domain orchestration, many proposals rely on heavy integration fabrics, multi-layer coordinators, and deep telemetry pipelines that hinder deployability and amplify coordination overhead. We propose a lightweight, decentralized conversational orchestration framework based on Large Language Model (LLM)-driven domain agents. Each domain remains autonomous: an agent observes local state via tools, reasons in a closed loop, and exchanges summaries with neighboring agents over an Agent-to-Agent (A2A) overlay aligned with data-plane coupling. Fast feasible placement is enabled by periodic, routing-like dissemination of reachability advertisements (latency, bottleneck bandwidth, and compute capacity), while safe re-optimization, scaling, and migration are handled through event-driven requests and negotiation. To meet real-time constraints, we deploy a compact reasoning model trained with verifier-based self-verification and periodically refined online via shadow updates. Simulations show manageable, near-linear control-plane overhead as domains scale and during domain joins, and robust decision quality, including recovery after objective changes. We close by outlining future research directions for principled, secure, and uncertainty-aware agentic orchestration in Organic 6G.
Tags
Links
- Source: https://arxiv.org/abs/2608.10714v1
- Canonical: https://arxiv.org/abs/2608.10714v1
Trouble viewing inline? Open PDF directly â
Full Text
42,845 characters extracted from source content.
Expand or collapse full text
1 Conversational Orchestration for Organic 6G Masoud Shokrnezhad 1 and Tarik Taleb 2 1 ICTFICIAL Oy, Espoo, Finland; masoud.shokrnezhad@ictficial.com 2 Ruhr University Bochum, Bochum, Germany; tarik.taleb@rub.de AbstractâThe Organic 6G vision of a network of networks spanning an edgeâcloud continuum complemented by non- terrestrial resources requires, to realize its promise, service provisioning that is simple to operate, scalable across indepen- dently administered domains, and agile under domain churn (i.e., domains dynamically joining and leaving). Despite advances in cross-domain orchestration, many proposals rely on heavy integration fabrics, multi-layer coordinators, and deep telemetry pipelines that hinder deployability and amplify coordination overhead. We propose a lightweight, decentralized conversa- tional orchestration framework based on Large Language Model (LLM)-driven domain agents. Each domain remains autonomous: an agent observes local state via tools, reasons in a closed loop, and exchanges summaries with neighboring agents over an Agent-to-Agent (A2A) overlay aligned with data-plane coupling. Fast feasible placement is enabled by periodic, routing-like dissemination of reachability advertisements (latency, bottleneck bandwidth, and compute capacity), while safe re-optimization, scaling, and migration are handled through event-driven requests and negotiation. To meet real-time constraints, we deploy a com- pact reasoning model trained with verifier-based self-verification and periodically refined online via shadow updates. Simulations show manageable, near-linear control-plane overhead as domains scale and during domain joins, and robust decision quality, including recovery after objective changes. We close by outlining future research directions for principled, secure, and uncertainty- aware agentic orchestration in Organic 6G. Index TermsâOrganic 6G, Service Provisioning, Multi- Domain Orchestration, Agentic AI, Large Language Models, Non-Terrestrial Networks. I. INTRODUCTION Organic 6G envisions the future mobile system as a software-centric network of networks, where heterogeneous radio, compute, and transport resources, spanning edge-to- cloud and including Non-Terrestrial Networks (NTN) assets, continuously appear, disappear, and reshape under independent administrative control [1]. This evolution aligns with the broader push toward open networks built on interoperable standards for cross-domain service delivery [2]. From a multi- domain perspective, this direction is not optional: end-to-end services inevitably traverse domains with distinct policies, trust boundaries, and operational constraints. Hence, 6G must be designed to function under decentralization and churn, supporting seamless domain onboarding/offboarding rather than assuming a fixed, single-operator topology. This structural dynamism (an infrastructure that must remain coherent and serviceable as resource domains continuously join, leave, and evolve) is the central challenge motivating this paper: any orchestration approach that assumes a static, globally known topology will fail in this setting. However, an Organic 6G infrastructure is only useful if services can be provisioned over it. In this paper, service provisioning denotes the end-to-end process of turning a service description into running, reachable service instances by (i) selecting and placing compute across the edgeâcloud con- tinuum and (i) establishing the required connectivity so users can be bound to those instances, while continuously adapting via scaling and migration as conditions change. Doing so across many independently administered domains requires methods that are scalable (bounded coordination cost as do- mains grow), simple (deployable and failure-resilient without heavy operational burden), and agile (plug-and-play domain join/leave, i.e., resource domains dynamically onboarding or going offline at runtime) [3]. While recent work has made progress toward cross-domain orchestration in 5G/6G settings, existing approaches often presume substantial architectural machinery (e.g., multi-layer coordinators, integration fabrics, deep telemetry/Artificial Intelligence (AI) pipelines) and pre- established federation agreements, which complicate plug-and- play dynamics and can drive coordination overhead upward with the number of domains [4]â[8]. To fill this gap, this paper proposes a lightweight, decen- tralized conversational orchestration approach based on Large Language Model (LLM)-driven domain agents that coordinate service provisioning decisions over an inter-domain overlay graph. The key insight is that LLM-based agents shift orches- tration complexity from engineered architectural machinery (integration fabrics, coordinator hierarchies, raw telemetry aggregation) to goal-driven reasoning. Each domain remains autonomous â its agent optimizes locally using domain tools and policies â yet end-to-end coordination emerges by ex- changing only goal-driven, summarized reachability informa- tion with neighboring agents via Agent-to-Agent (A2A) mes- saging. Concretely, the proposed design combines (i) periodic, routing-like dissemination of compact resource reachability advertisements and (i) on-demand request/negotiation for safe re-optimization and migration, thereby keeping coordination bounded while supporting domain churn. This keeps the design scalable (localized exchanges, no centralized entity), simple (no heavy orchestration machinery), and agile (plug-and-play domain join/leave). The remainder of this paper is organized as follows: Sec- tion I provides background and requirements; Section I presents the proposed agent-based architecture and decision process; Section IV evaluates the approach via simulations; and Section V discusses future directions and concludes this paper. I. BACKGROUND A. Infrastructure From a resource view, Organic 6G treats the infrastructure as a distributed continuum of connectivity and computation rather than a fixed, centralized stack. At the access edge, the Radio Access Network (RAN) (including disaggregated/vRAN arXiv:2608.10714v1 [cs.NI] 11 Aug 2026 2 oRAN Near-Edge Regional Micro-DCCentral Cloud Terrestrial \ Underground Aerial Space Far-Edge vRAN Domains= Administrative Boundaries Links = Heterogeneous x-haul/NTN Fig. 1. Organic 6G infrastructure as a distributed continuum of connectivity and computation spanning terrestrial, aerial, and space assets from near-edge to central cloud, organized as a modular network of networks with isolated administrative domains interconnected via heterogeneous x-haul/NTN. and Open RAN components) anchors the system by attaching User Equipment (UEs) and exposing radio resources that can be shaped per service. Beyond the RAN, computation spans in-network elements, near-edge nodes (co-located with or adjacent to the RAN), far-edge nodes (district/metro- level aggregation points), regional micro-data centers, and central clouds, forming a continuum that trades latency for capacity. The interconnect is inherently heterogeneous: front- /mid-/backhaul spans fiber, copper, terrestrial wireless, and NTN assets such as satellites, Unmanned Aerial Vehicles (UAVs), and High Altitude Platforms (HAPs)/Low Altitude Platforms (LAPs), each with distinct bandwidth, latency, and intermittency profiles [9]; this heterogeneity extends reach and resilience where terrestrial links are sparse or disrupted. Hardware heterogeneity (Field-Programmable Gate Arrays (FPGAs), Smart Network Interface Cards (SmartNICs), GPUs) further supports efficient data-plane execution at the edge. At a system level, Organic 6G interprets this continuum as a modular network of networks built from isolated administrative domains, where each domain can own and operate a well- scoped subset of the overall resource pool. In practice, one entity may provide the access segment (e.g., a local RAN deployment), another may operate a portion of the com- pute continuum (e.g., near- or far-edge nodes, or regional micro-data centers), while additional entities may contribute transport/x-haul capacity or non-terrestrial resources. Each do- main encapsulates its resources behind domain-local policies and management functions, enabling independent lifecycle management without assuming a static, globally engineered topology. The result is an infrastructure that remains struc- turally coherent even when its constituent resource domains appear, disappear, or change shape over time. The co-existence of these heterogeneous resources, each operated by an inde- pendent administrative domain with its own policies and ca- pabilities, is what makes end-to-end service provisioning non- trivial: a single service may require compute at multiple tiers, connectivity across heterogeneous link types, and continuous adaptation as users move or conditions change, all without any single entity having global visibility or control. Fig. 1 provides an overview of the above. Unlike ephemeral networks (temporary by design) or end-to-end network slicing, which virtualizes resources over a globally orchestrated substrate, Organic 6G describes a persistent, multi-domain network of networks where the physical infrastructure itself evolves dynamically under independent administrative control, with no single entity holding a global view. B. Service Provisioning Building on the infrastructure continuum described above, service provisioning refers to the end-to-end process of turning a service description into running, reachable service instances over the available computing and networking resources. For example, a latency-sensitive service may be placed on a near- edge node in the userâs current domain, with connectivity established through the local RAN and x-haul; as the user moves across domain boundaries, the instance is migrated to a near-edge node in the new domain to preserve Quality of Service (QoS). Concretely, it entails placing service instances across the edgeâcloud spectrum and establishing RAN/x-haul connectivity to bind users to those instances [10]. In Organic 6G, provisioning is driven by stringent QoS objectives under user mobility and highly dynamic resource conditions: latency is often the dominant metric for interactive and control work- loads, while availability captures whether sufficient compute and connectivity resources remain continuously accessible to sustain a userâs service experience. Beyond initial placement, service provisioning includes life- cycle adaptation operations that keep the service performant as conditions change. Scaling adjusts the number of active instances and their allocated resources to follow demand and to mitigate hotspots, whereas migration relocates an instance (or its execution context) to a different node to preserve QoS as users move or as local resources become scarce or unreliable. Together, placement, binding, scaling, and migration form the core of service provisioning in Organic 6G: because the environment is highly dynamic (user loads shift, users move, and infrastructure conditions fluctuate), a static one- time placement is insufficient, and continuous adaptation is required to maintain a valid mapping between users, service instances, and infrastructure resources so that QoS objectives remain satisfied. Section I describes how the proposed agent- based approach addresses these challenges through continuous reachability tracking, event-driven re-optimization, and online model refinement. C. Problem Statement & Challenges The core problem addressed in this paper is the optimization of service provisioning over an Organic 6G infrastructure. Given the distributed pool of computing resources (from near- edge to central and non-terrestrial nodes) and networking resources (the RAN and heterogeneous transport/x-haul), an orchestrator must (i) collect and maintain sufficiently accurate system information across the network of networks, (i) de- cide provisioning actions, including placement, userâinstance binding, scaling, and migration, and (i) enforce these actions 3 on the underlying resources. This closed-loop must operate under stringent QoS requirements (e.g., latency) while the system exhibits high dynamism. Importantly, any practical solution must satisfy three system-level requirements beyond single-domain optimality [3]: scalability (coordination cost must remain bounded as domains grow, avoiding centralized bottlenecks and single points of failure); simplicity (the ar- chitecture must be deployable and failure-resilient without heavy operational burden); and agility (domains must support plug-and-play onboarding/offboarding so the system remains functional as resource domains appear, disappear, or change shape). Recent work has investigated cross-domain orchestration for service provisioning in multi-domain 5G/6G environments. Giannopoulos et al. [4] proposed ACROSS, a two-level ar- chitecture where domain orchestrators are supervised by a cross-domain coordinator through a standardized integration fabric, relying on deep telemetry and AI-assisted zero-touch provisioning automation. Molner et al. [5] introduced AIORA, advocating virtual continuums spanning multiple segments and a cross-segment coordination substrate built atop open interfaces and nested AI-driven closed loops. Dalgitsis et al. [6] addressed inter-operator continuity via cloud-native slice federation, translating slice templates into federated templates exchanged over operator federation interfaces to instantiate slices in visited domains. Benlloch-Caballero et al. [7] stud- ied end-to-end slicing for security automation, coupling dis- tributed sensing and topology tracking with an orchestration loop and cross-domain enforcement agents. Finally, Santos et al. [8] proposed a hierarchical orchestration scheme for end-to-end network slicing, introducing a higher-level hy- perstrator to coordinate distributed per-segment orchestrators across domains, while relying on pre-defined inter-orchestrator interfaces and a central coordination point for lifecycle man- agement. While the achievements are valuable, they still fall short of the simplicity, agility, and scalability requirements emphasized above: they typically presume substantial archi- tectural machinery (integration fabrics, multiple orchestration layers, deep telemetry/AI pipelines, or custom enforcement components), pre-established interoperability and federation agreements that complicate plug-and-play domain join/leave, and coordination patterns whose overhead exponentialy grows with the number of domains and slices (e.g., centralized supervision, heavy telemetry aggregation, or increasing inter- orchestrator messaging). These gaps motivate our lightweight, decentralized agent- based coordination approach in Section I, designed to keep coordination bounded, operational complexity low, and do- main dynamics seamless at Organic 6G scale. I. PROPOSED APPROACH A. Architecture We build our solution around LLM-based agents: au- tonomous control entities that observe their environment, rea- son, and act in closed loop. LLMs are chosen over analytical or pure data-driven alternatives because their goal-driven reason- ing generalizes across heterogeneous domain configurations Remembering LLM Memory Short- Term Long- Term retrieve History / state store outcomes / feedback Resource Tools Peer - Agent Tools Tools tool calls/ actions live observations User Operator Other Agents provisioning request system prompt (objective, policies, ...) neighbor inquiries (A2A) Adapter Modern Software- Controlled Resources Legacy Infrastructure NMS / Cloud-Edge Controller unifiedtool interface (MCP, ...) local state summary placement / binding / scaling / migration Domain A Agent A Agent B Agent C Agent D Agent E A2A Fig. 2. Illustration of the proposed LLM-based domain agent and its inter- domain overlay control plane for Organic 6G service provisioning. The LLM is the reasoning core; memory provides working context and long- term history; the tool layer interfaces with domain resources and peer agents via A2A; and the adapter bridges the agent to both modern software-defined infrastructure (e.g., via MCP) and legacy NMS/API stacks. and adapts to objective changes without re-engineering (the accompanying Reinforcement Learning (RL)-based special- ization then ensures task-specific optimization competence). Here, the LLM serves as the agentâs reasoning core with five internal capabilities: goal setting (what to optimize), decision synthesis (generating candidate actions), memory (storing and retrieving state and feedback), action execution (enforcing decisions via tools), and self-correction (revising the reasoning path based on outcomes). These capabilities are realized with three modules: the LLM itself, a memory store (short-term working context plus persistent long-term history), and a tool layer that interfaces with resources and peer agents. An overview of the proposed agent architecture is shown in Fig 2. This design is deliberately goal-oriented: rather than relying on predefined orchestration pipelines, the agent selects which observations to request, which tools to invoke, and which control actions to apply based on the currently declared goals and constraints. The claimed system properties (simplicity, agility, scalability) are architectural properties of the orches- tration system, not of the model: the modelâs computational cost is local to each domain and does not add cross-domain coordination overhead, and Section I-C describes the use of a compact Small Language Model (SLM) to keep per-domain inference latency low. From an intra-domain point of view, we place one agent at the top of each administrative resource domain, responsible for domain-local resource allocation in support of service provisioning. The domain agent continuously collects and summarizes local state (e.g., available compute/network ca- pacity, current allocations, local policies, and observed QoS), receives provisioning requests, and enforces actions such as placement, userâinstance binding, scaling, and migration. It can also revise existing allocations to remain optimized as de- mand and infrastructure conditions change. The agent connects to the domain through an adapter layer: for modern, software- controlled resources it can interact directly via a unified tool interface (e.g., Model Context Protocol (MCP) connectors 4 toward devices/controllers), while for legacy infrastructures it operates through existing management stacks such as NMSs or cloud/edge controllers that expose monitoring and actuation APIs. From an inter-domain point of view, the agents form an overlay control-plane graph, where each node corresponds to a domain agent and each edge represents a communication relationship [11]. We align this overlay with data-plane cou- pling: if two domains are connected (or interdependent) in the data plane, their agents are connected in the control plane and can exchange information about relevant cross-domain state. Importantly, by placing the coordination logic in the agents, domains are not required to implement new cross-domain co- ordination standards; instead, agents communicate using A2A protocols on the control plane. As a result, the architecture directly improves: scalability by limiting coordination to goal- driven, neighbor-to-neighbor exchanges rather than raw global state collection; simplicity by shifting orchestration logic from fragile, manually engineered workflows to an agentic closed loop over tools and feedback; and agility by supporting plug- and-play join/leave at the domain level (a domain onboards by deploying its agent+adapters and establishing edges to adjacent domains). B. Decision Making Our agents coordinate cross-domain provisioning using two complementary message-passing modes on the overlay control plane. The first is change-triggered dissemination, which maintains a routing-like view of feasible resources; the second is on-demand request/negotiation, which supports safe re-optimization and migration. In both cases, each domain agent remains authoritative only within its own administrative boundary: it advertises summarized reachability information, but enforces allocations (admission, placement, binding, scal- ing, migration) solely through its domain-local tool layer and policies. An overview is illustrated in Fig. 3. Change-triggered dissemination (table-driven). A domain agent summarizes the feasibility of reaching its local comput- ing resources from its data-plane inter-domain ingress/egress points. Concretely, for each reachable compute resource it derives a compact resource advertisement combining end-to- end latency, path bottleneck bandwidth, and available compute capacity. The agent then shares this advertisement with neigh- boring agents over A2A. Upon receipt, a neighbor performs a routing-style update: for each advertised destination it adds its own intra-domain access latency to the destination, tightens the available bandwidth to reflect the new bottleneck, and records the next-hop neighbor toward that destination. This process repeats whenever a domain agent observes a material change in its intra-domain resources or connectivity (or receives an updated advertisement from a neighbor that changes its table). By repeating these local updates, agents build and refresh a distributed resource reachability table that maps feasible remote computing resources to their predicted end-to-end latency and available capacities, without requiring raw global telemetry collection. Request-driven provisioning (per-request). When a pro- visioning request arrives at a domain, the receiving agent first fetch table update table UserNeighbor Agent LLM Memory Resource a change in resources calculate table request a service reason for placement / binding. any possible allocation? (y) implement solution (n) reject request fetch state change in neighborâs table (y) update history reason for migration / scaling. any possible allocation? (y) implement solution Negotiate for migration fetch state (y) update history request accepted return service state fetch service state distribute service state Update Resource Reachability Table If There Is No Local Solution Service Provisioning Requested by Another Domain On-demand Intra-Domain Resource Reallocation Domain Agent A2A Service Provisioning Requested by a User request a service reason for placement / binding. any possible allocation? (y) implement solution fetch state update history (n) fetch table select a domain distribute the request request rejected? (y) reject request Disseminate update / advertisement Dissemination Request - Driven Provisioning Fig. 3. High-level procedures for change-triggered reachability dissemination and service provisioning on the inter-domain agent overlay. For tractability, the figure omits detailed per-hop mechanisms (e.g., multi-hop forwarding, soft reservations/commit). checks whether it can satisfy the request locally (subject to domain policies and current allocations). Otherwise, it consults its reachability table to select a feasible remote destination whose advertised compute and bandwidth capacities satisfy the request and whose predicted end-to-end latency meets the requestâs QoS constraints; among candidates, it prefers the one minimizing latency. The request is then forwarded hop-by-hop on the control plane until it reaches the destination domain agent. Each intermediate agent installs a soft reservation on its segment and the outbound inter-domain link; once the destination domain allocates compute, a confirmation travels 5 back and each domain commits its reservation, binding the control-plane chain to an end-to-end data-plane path while preserving domain autonomy. Because feasibility information is already available in the table and reservations are estab- lished incrementally per hop, this mode supports low decision latency for per-request allocations while keeping coordination bounded to neighbor exchanges. On-demand negotiation (event-driven). Disseminated ta- bles are intentionally coarse: they enable fast, feasible place- ment, but they do not by themselves guarantee that subsequent changes (e.g., intra-domain path rebalancing, resource reallo- cation, or migration) preserve each serviceâs QoS. For these cases, agents switch to an on-demand mode in which an agent initiating a change (e.g., flow rerouting, migration) negotiates with the agent(s) responsible for the affected service in- stance(s) via A2A messaging, exchanging the additional state needed (updated path characteristics, migration context) to confirm that end-to-end QoS is preserved before enforcement. While such negotiation introduces additional control latency, it is used for optimization and lifecycle adaptation rather than the critical path of initial admission, thereby enabling improved optimality without imposing a centralized orchestrator. C. Training & Inference To operate as domain-top controllers, the proposed agents must generate provisioning decisions that are fast yet suf- ficiently intelligent for multi-constraint optimization. Ac- cordingly, each domain agent is instantiated with an SLM pretrained on strong reasoning tasks as its reasoning core (DeepSeek-R1-Distill-Qwen-7B in the evaluation): this choice preserves low inference latency while providing the reasoning competence needed for reliable, goal-driven control (training procedure and evaluation are detailed in what follows and in Section IV). However, service provisioning in Organic 6G requires more than generic reasoning: the model must internalize domain policies, optimize under multiple criteria (e.g., latency and resource efficiency), and produce actionable allocations that satisfy QoS constraints. To this end, the SLM is specialized using RL with verifiable feedback signals, following the self-verification paradigm popularized by recent reasoning models [12]. Offline self-verification training. Fig. 4 summarizes the offline self-verification procedure. A set of training contexts is constructed to reflect the foundations in Section I and the decision process in this section: each context encodes (i) a domain view of compute and network resources, (i) local poli- cies and constraints, (i) representative provisioning requests, and (iv) target optimization criteria. For each context, the SLM generates a reasoning trace and a candidate provisioning action (placement, userâinstance binding, scaling, or migration). A stronger verifier LLM then assesses the output and assigns a multi-objective reward vector capturing three aspects: (1) reasoning validity (is the allocation supported by a coherent and complete reasoning path, avoiding unjustified shortcuts [13]), (2) optimization quality (how well the proposed decision matches the declared objectives such as latency minimization under capacity constraints), and (3) QoS feasibility (whether Training Contexts (domain snapshots) â˘Resourcesview (compute+ network) â˘Policies& constraints â˘Requests(placement/ scaling / migration) â˘Objectives (latency, efficiency, etc.) Domain SLM Contexts Provisioning action / Reasoning trace GDPO Update Rewards: Reasoning Validity / Optimization quality / QoS feasibility Update Meta Verifier Verifier LLM Contexts Contexts Provisioning action / Reasoning trace few-shot promptupdates Live Decision Episodes â˘Live resourcestate(tools/telemetry) â˘Currentrequest â˘Domain policies/objectives(prompt) â˘Retrievedmemory/history Active SLM Live episode Reasoning trace Shadow GDPO Self-score: Reasoning Validity / Optimization Quality / QoS feasibility Update Self Verifier (SLM) Shadow SLM Domain Adapter Infrastructure Provisioning action Live episode Trace log Observed QoS Live episode Training Batch Periodic replace Fig. 4. Offline self-verification training of domain SLMs. Training contexts (domain snapshots) elicit reasoning traces and provisioning actions; a strong verifier assigns multi-objective rewards (reasoning validity, optimization qual- ity, and QoS feasibility) that drive GDPO updates. Verifier reliability is improved via meta-verifierâguided few-shot prompt refinement. Training Contexts (domain snapshots) â˘Resourcesview (compute+ network) â˘Policies& constraints â˘Requests(placement/ scaling / migration) â˘Objectives (latency, efficiency, etc.) Domain SLM Contexts Provisioning action / Reasoning trace GDPO Update Rewards: Reasoning Validity / Optimization quality / QoS feasibility Update Meta Verifier Verifier LLM Contexts Contexts Provisioning action / Reasoning trace few-shot promptupdates Live Decision Episodes â˘Live resourcestate(tools/telemetry) â˘Currentrequest â˘Domain policies/objectives(prompt) â˘Retrievedmemory/history Active SLM Live episode Reasoning trace Shadow GDPO Self-score: Reasoning Validity / Optimization Quality / QoS feasibility Update Self Verifier (SLM) Shadow SLM Domain Adapter Infrastructure Provisioning action Live episode Trace log Observed QoS Live episode Training Batch Periodic replace Fig. 5. Online inference and periodic refinement with shadow updates. The deployed SLM agent retrieves memory, queries live state via tools, enforces provisioning actions through the domain adapter, and logs traces together with observed QoS. Traces are batched to update a shadow copy offline (e.g., GDPO), which periodically replaces the deployed model. the requests can be served within their QoS bounds given the advertised resources). These heterogeneous rewards are used to update the SLM with Group reward-Decoupled Normalization Policy Optimization (GDPO), which stabilizes training by nor- malizing each reward component before aggregation, thereby preserving learning signal resolution in multi-reward settings [14]. In parallel, the verifier is improved through a lightweight bootstrapping loop: a meta-verifier reviews verifier judgments, and the resulting feedback is injected into the verifier prompt (few-shot style) to progressively reduce systematic errors, without requiring additional training infrastructure. Online inference and periodic refinement. Fig. 5 il- lustrates the online closed loop and the periodic shadow- update refinement strategy. After offline training, the SLM- based agent is deployed on top of each domain with access to its memory and tool layer, and with a system prompt encoding the domainâs policies and optimization criteria. At inference time, the agent follows the closed loop described earlier: it retrieves relevant history, collects live state via 6 tools, proposes a decision, and enforces it through the domain adapter. Reliability is improved by reusing the same self- verification loop as in offline training, which also serves as a lightweight hallucination guard: outputs violating QoS feasibility or lacking coherent reasoning are penalized before enforcement. For each decision, the SLM produces a reasoning trace and an allocation, then evaluates its own output using the same verifiable, multi-objective reward criteria learned during training. To adapt to evolving traffic and infrastructure conditions without blocking real-time control, inference traces (requests, state snapshots, actions, rewards, and observed QoS) are accumulated into batches, and a separate copy of the SLM is updated periodically offline (e.g., with GDPO). From time to time, the updated copy replaces the working SLM [15]. This shadow update strategy decouples live decision making from continual retraining, while enabling steady improvement under non-stationary Organic 6G dynamics. IV. SIMULATIONS A. Scenario A: Control-Plane Evaluation We quantify control-plane message volume to assess the overhead of the dissemination in Section I, focusing on (i) reachability-table formation and (i) re-convergence after a domain joins. The results are illustrated in Fig. 6.A. We simulate N â 10, 20, 30 administrative domains whose agents form an overlay aligned with data-plane coupling. Each curve is averaged over 20 independent runs. In each run, N domains (each exposing 3 local compute resources summarized by its agent) are connected by a random inter- domain graph with target average degree 4, and the agents execute table-driven dissemination. Each time slot represents one management epoch (consistent with orchestration decision intervals on the order of seconds), so convergence within tens of slots corresponds to table formation on the order of minutes, appropriate for provisioning-level control. When a local table change occurs, the agentâs message-preparation latency is mea- sured and then normalized to a 1â3 time-slot interval before the update is emitted to neighbors; delivery succeeds with probability p = 0.9 and reliability is enforced by ACK-based retransmissions. To match the accounting in Fig. 6.A, each changed table entry is treated as a separate update message (sent in parallel within the same time slot), and we report update messages (ADV) only. The initial burst occurs while agents rapidly populate reachability tables with previously unknown destinations; once tables stabilize, change-triggered updates cease and the message rate collapses to near zero. At time slot 60, a new domain joins and attaches to a ran- dom subset of existing domains, introducing only incremental reachability entries, which yields a smaller transient before re- convergence. Overall, the overhead remains manageable, and for fixed average degree it increases approximately linearly with the number of domains, as more destinations must be disseminated over neighbor-only exchanges. B. Scenario B: Training and Refinement Evaluation This scenario validates the two key claims of Section I-C: that offline RL specialization brings the SLM to near-verifier- level provisioning performance, and that online refinement objective switch A B Fig. 6. Simulation results for the two scenarios. (A) Scenario A: Message- passing overhead of distributed reachability-table dissemination, including a domain join at time slot 60. Lines show the mean number of update messages per time slot; shaded regions denote minâmax across random topologies for eachN . (B) Scenario B: Normalized score (%) relative to DeepSeek- R1 (verifier LLM baseline fixed at 100%). The SLM improves with offline specialization on load-balance (epochs 0â50); after switching to a min-latency objective (epochs 50â100), only the variant with online refinement recovers toward the baseline. enables recovery under objective changes. We evaluate the offline and online specialization loop using DeepSeek-R1 as the strong verifier LLM and DeepSeek-R1-Distill-Qwen- 7B as the deployed SLM reasoning core, over a synthetic heterogeneous multi-domain infrastructure whose parameters (12 domains, inter-domain latencies 2â20 ms, bandwidths 0.5â10 Gbps, and per-domain compute pools spanning 32â 128 vCPU-equivalents) span the plausible operating range of 6G deployments. We generate a dataset of 3000 service- provisioning scenarios over this infrastructure. Training pro- ceeds in two phases. During epochs 0â50, the SLM is spe- cialized offline toward a load-balance objective (balancing resource utilization). From epoch 50 onward, the objective switches to min-latency (minimizing overall provisioning la- tency). Both phases are also subject to QoS and capacity feasi- bility. We evaluate at checkpoints every 10 epochs and report a normalized score under the active objective, where DeepSeek- R1 is used as the 100% reference. Fig. 6.B compares three variants: (i) a base SLM without training, (i) SLM with offline training (trained for load-balance only), and (i) SLM with offline training followed by online refinement via periodic shadow updates. Offline training drives the SLM close to the DeepSeek-R1 reference on the load-balance objective. After the objective switch, the offline-only variant drops and remains lower, whereas online refinement restores performance as the model adapts to the new min-latency criterion. Overall, the results indicate that the proposed offline+online specializa- tion yields sufficient capability to maintain high accuracy 7 relative to the verifier LLM under both stationary operation and objective changes. Practically, achieving near-verifier-level scores on the min-latency objective means the deployed SLM produces placement and binding decisions that approach the quality of a full LLM, translating directly to tighter end- to-end service latency for users while keeping inference lo- cal to each domain. Note that Scenario A and Scenario B evaluate complementary aspects: the former assesses control- plane dissemination overhead (independent of the decision- maker), while the latter evaluates SLM reasoning quality and adaptation. V. CONCLUSION & FUTURE WORK In this paper, we studied service provisioning over an Organic 6G infrastructure by viewing the system as a het- erogeneous edgeâcloudâNTN continuum composed of inde- pendently administered resource domains. We formulated the problem as a continuous placement-and-binding loop under stringent QoS targets, subject to scalability, simplicity, and agility requirements. To address these, we proposed a de- centralized architecture based on LLM-based domain agents connected by an A2A overlay graph: agents used table- driven dissemination to build reachability views for fast fea- sible placement, and switched to event-driven negotiation for safe re-optimization and migration. Simulations validated that neighbor-only dissemination induced manageable control- plane overhead, and that an SLM reasoning core specialized via verifier-based self-verification and periodic refinement maintained high decision quality, including under objective changes. Looking forward, we outline several fundamental research directions for making agentic orchestration a principled and robust foundation for Organic 6G: ⢠Formal cross-domain agentic orchestration: develop formal models for cross-domain trade-offs and scalabil- ity limits (including domain-count thresholds at given performance targets); current simulations demonstrate feasibility up to 30 domains with near-linear overhead, but formal bounds remain open. ⢠Uncertainty-aware decision theory: design methods that represent uncertainty and provide tail-risk QoS guar- antees under non-stationary dynamics (e.g., fluctuating link latency or intermittent compute availability at near- edge nodes). ⢠Security for agentic control planes: establish threat models and defenses for A2A-mediated orchestration (e.g., a malicious domain agent injecting false reachabil- ity advertisements to attract or divert traffic, confused- deputy behavior, or compromised tool responses), to- gether with adversarial benchmarks. ⢠Multi-agent stability and conflict resolution: study agent-population conflicts and oscillations (e.g., two neighboring agents repeatedly migrating the same service instance back and forth, or a cascade of reallocations trig- gered by a single domain change), and design lightweight arbitration and convergence mechanisms. ⢠Mechanism design for truthful coordination: study incentive-compatible protocols for truthful resource ad- vertisement (i.e., without incentive to overstate or un- derstate available resources for strategic gain) and fair sharing, aligning local utilities with end-to-end QoS. ⢠Learning and adaptation under real-time constraints: quantify adaptabilityâcost trade-offs: safe online learning, energy/latency budgets, and privacy-preserving or feder- ated adaptation in heterogeneous environments. ACKNOLWEDGEMENT This work was, in part, supported by BMFTR, Germany (6GEM+, Grant 16KIS2411), and the EU Horizon Europe programme (6G-Path, Grant 101139172). REFERENCES [1] M. I. Corici, F. Eichhorn, R. Bless et al., âOrganic 6G Networks: Vision, Requirements, and Research Approaches,â IEEE Access, vol. 11, p. 70 698â70 715, 2023. [2] S. Ghosh, H. Asgari, D. Hond et al., âFuture Open Networks Cross- Domain Cognitive Orchestration: A Novel Design Paradigm,â IEEE Access, vol. 13, p. 105 911â105 951, 2025. [3] âHorizon Europe Call for Advanced Architectures Systems and Technologies,â Accessed: Feb. 3, 2026. [Online]. Available: https: //tinyurl.com/2wyjkap6 [4] D. Giannopoulos, G. Katsikas, K. Trantzas et al., âACROSS: Automated zero-touch cross-layer provisioning framework for 5G and beyond vertical services,â in 2023 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit), Jun. 2023, p. 735â740, iSSN: 2575-4912. [5] N. Molner, L. Rosa, F. Risso et al., âAIORA: An AI-Native Multi- Stakeholder Orchestration Architecture for 6G Continuum,â IEEE Net- work, p. 1â1, 2025. [6] M. Dalgitsis, N. Cadenelli, M. A. Serrano et al., âCloud-Native Orches- tration Framework for Network Slice Federation Across Administrative Domains in 5G/6G Mobile Networks,â IEEE Transactions on Vehicular Technology, vol. 73, no. 7, p. 9306â9319, Jul. 2024. [7] P. Benlloch-Caballero, A. Matencio-Escolar, J. Bernal Bernabe et al., âE2E Network Slicing for Enhanced Cybersecurity, Orchestration, Au- tomation and Response in 5G/6G: The RIGOUROUS Approach,â Jour- nal of Network and Systems Management, vol. 34, no. 1, p. 22, Nov. 2025. [8] J. F. Santos, W. Liu, X. Jiao et al., âBreaking down network slicing: Hi- erarchical orchestration of end-to-end networks,â IEEE Communications Magazine, vol. 58, no. 10, p. 16â22, 2020. [9] M. Shokrnezhad, H. Yu, T. Taleb et al., âToward a Dynamic Future With Adaptable Computing and Network Convergence (ACNC),â IEEE Network, vol. 39, no. 2, p. 268â277, Mar. 2025. [10] M. Farhoudi, M. Shokrnezhad, and T. Taleb, âService Registration, Indexing, Discovery, and Selection: An Architectural Survey Toward a GenAI-Driven Future,â IEEE Access, vol. 13, p. 209 680â209 722, 2025. [11] C. Wittner, âCommunication Methods in Multi-Agent Reinforcement Learning,â Jan. 2026, arXiv:2601.12886. [Online]. Available: http: //arxiv.org/abs/2601.12886 [12] DeepSeek-AI, D. Guo, D. Yang et al., âDeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning,â Jan. 2025, arXiv:2501.12948. [Online]. Available: http://arxiv.org/abs/2501.12948 [13] T. Xie, Z. Gao, Q. Ren et al., âLogic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning,â Feb. 2025, arXiv:2502.14768. [Online]. Available: http://arxiv.org/abs/2502.14768 [14] S.-Y. Liu, X. Dong, X. Lu et al., âGDPO: Group reward- Decoupled Normalization Policy Optimization for Multi-reward RL Optimization,â Jan. 2026, arXiv:2601.05242. [Online]. Available: http://arxiv.org/abs/2601.05242 [15] Q.-A. Dang and C. Ngo, âReinforcement Learning for Reasoning in Small LLMs: What Works and What Doesnât,â Mar. 2025, arXiv:2503.16219. [Online]. Available: http://arxiv.org/abs/2503.16219