|
1.
|
Clear description of the referenced document:
|
|
|
|
Name:
|
ETSI GR ENI 056 V4.1.1
|
|
Title:
|
Experiential Networked Intelligence (ENI); Study on Multi-Agent Frameworks for Next-Generation Core Networks
|
|
|
2.
|
Status of approval:
|
|
|
Published ETSI Group Report, produced and approved by the members of ETSI ISG ENI. Version V4.1.1 (October 2025) is the current publicly available edition; according to the document history recorded in the GR, this is the first published (“Publication”) milestone of this deliverable.
|
|
3.
|
Justification for the specific reference:
|
|
|
In F.CV-AIVVA, the task planner is described as being required to acquire rich expert knowledge and network information by repeatedly coordinating interactions with external multiple agent services. ETSI GR ENI 056 V4.1.1 (2025-10) is referenced because it defines, for a comparable multi-agent context, the same category of coordination behaviour and provides the architectural basis and terminology on which this description relies.
Specifically, clause 6.2.3 (Agent-to-Agent) of the referenced document describes a Planning Agent that decomposes a user intent into sub-tasks and repeatedly invokes other specialized agents (e.g. an Assemble Agent) via a task-centric interface; this is directly analogous to the task planner’s role of coordinating with external agent services in F.CV-AIVVA. Clause 6.2.4 (Agent-to-Resource) describes how agents access external data services, tool repositories and knowledge bases to obtain the information needed for a task, which corresponds to the acquisition of expert knowledge and network information referred to in the task planner requirement.
Clause 5 (Multi-Agents System in Core Network) of the referenced document also motivates, at the architectural level, why a task requiring broad expert knowledge is better served by coordinating with multiple specialized external agents rather than by a single generalized agent, which supports the design rationale underlying the task planner architecture adopted in F.CV-AIVVA.
|
|
4.
|
Current information, if any, about IPR issues:
|
|
|
No known issues.
|
|
5.
|
Other useful information describing the "Quality" of the document:
|
|
|
ETSI GR ENI 056 V4.1.1 is produced by ETSI ISG ENI following the ETSI Drafting Rules and the standard ETSI ISG review and approval process. Its technical content is grounded in a review of established open-source multi-agent frameworks, published protocol specifications (e.g. A2A, MCP, ACP, ANP) and arXiv research literature, and includes concrete interface definitions, message examples, and comparison tables of existing frameworks and protocols.
|
|
6.
|
The degree of stability or maturity of the document:
|
|
|
ETSI Group Reports (GR) are informative study deliverables rather than conformance-tested normative specifications (the latter being published by ETSI as Group Specifications, GS). V4.1.1 (October 2025) is the first published edition of this GR. The multi-agent architecture, interface taxonomy and collaboration mechanisms relevant to the reference are already stable and published. ETSI ISG ENI has an established track record of related, previously published deliverables (e.g. ETSI GR ENI 016, ETSI GS ENI 005, ETSI GS ENI 019, ETSI GR ENI 051), which provides continuity for this GR.
|
|
7.
|
Relationship with other existing or emerging documents:
|
|
|
ETSI GR ENI 056 V4.1.1 focuses on applying Multi-Agent Systems (MAS) within next-generation core networks, exploring how to coordinate multiple AI agents to solve complex tasks collaboratively. Conversely, other standards like Recommendation ITU-T F.748.46 primarily focus on the core capabilities of individual agents. The latter provides a foundational framework of requirements and evaluation methods for large-model-driven single or general AI agents, while the former studies how to apply these agents within network architectures for collaborative optimization.
|
|
8.
|
Any explicit references within that referenced document should also be listed:
|
|
|
Section 2) Normative references, added:
[1] 3GPP TS 23.501 (V19.2.1): "System architecture for the 5G System (5GS)".
[2] 3GPP TS 29.500 (V19.2.0): "Technical Realization of Service Based Architecture".
[3] 3GPP TS 29.501 (V18.3.0): "Principles and Guidelines for Services Definition".
[4] IEEE 802.11™: "IEEE Standard for Information Technology––Telecommunications and Information Exchange between Systems Local and Metropolitan Area Networks––Specific Requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications".
[5] 3GPP TS 33.501 (V19.1.0): "Security architecture and procedures for 5G system".
[6] J. Kaplan, et al.: "Scaling Laws for Neural Language Models", arXiv preprint arXiv:2001.08361, 2020.
[7] Q. Wu, et al.: "AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversations", Conference on Language Modeling, 2024.
[8] T. Guo, et al.: "Large Language Model Based Multi-Agents: A survey of progress and challenges", arXiv preprint arXiv:2402.01680, 2024.
[9] C. Chan, et al.: "Chateval: Towards better LLM-based Evaluators Through Multi-Agent Debate", arXiv preprint arXiv:2308.07201, 2023.
[10] GitHub repository for the Agent Network Protocol.
[11] J. Rosenberg, C. Jennings: "Framework, Use Cases and Requirements for AI Agents" (accessed on 08 May 2025).
[12] ETSI GR ENI 051 (V4.1.1): "Experiential Networked Intelligence (ENI); Study on AI Agents based Next-generation Network Slicing".
[13] ETSI GR ENI 016 (07-2021): "Experiential Networked Intelligence (ENI); Functional Concepts for Modular System Operation".
[14] ETSI GS ENI 019 (V4.1.1): "Experiential Networked Intelligence (ENI); Representing, Inferring, and Proving Knowledge in ENI".
[15] Agent Communication Protocol.
[16] Anthropic: "Building Effective Agents", 2024.
[17] A. Ehtesham, A. Singh, G. K. Gupta, S. Kumar: "A Survey of Agent Interoperability Protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)", arXiv preprint arXiv:2505.02279, 2025.
[18] Y. Lu, J. Wang: "KARMA: Leveraging Multi-Agent LLMs for Automated Knowledge Graph Enrichment", arXiv preprint arXiv:2502.06472, 2025.
[19] ETSI GS ENI 005 (V4.1.1): "Experiential Networked Intelligence (ENI); ENI System Architecture".
[20] Model Context Protocol (MCP).
[21] A2A Protocol.
[22] Semantic Versioning.
|
|
9.
|
Qualification of
ETSI:
|
|
|
ETSI is recognized under the provisions of ITU-T Recommendations A.5 and A.6. Qualifying information is on file at TSB.
|
|
10.
|
Other (for any supplementary information):
|
|
|
None
|
|
