arXiv：cs.AI· Taotao Wang, Lizhao You, Jingwen Tong, Chonghe Zhao, Shengli Zhang·· 13 小时前
Agentic P2P 网络：从内容分发到能力与行动共享
Agentic Peer-to-Peer Networks: From Content Distribution to Capability and Action Sharing
AI 导读
论文提出 Agentic Peer-to-Peer（P2P）网络，用于客户端自主智能体（CSAA）之间交换动态、异构的能力而非静态内容。作者给出分层参考架构，将连接、语义发现与执行解耦，并引入签名软状态能力描述符支持意图感知路由。为应对对抗环境，论文提出从声誉信号、canary 探针到密码学证据的三级验证谱系，代理级离散事件仿真显示在投毒和漂移下工作流成功率提升，控制面开销适中。
正文
Abstract:The ongoing shift of Artificial Intelligence (AI) from centralized cloud application programming interfaces (APIs) to local inference on edge devices is enabling Client-Side Autonomous Agents (CSAAs) - persistent personal agents that can plan, access local context, and invoke tools on behalf of users. As these agents collaborate directly, they form Agentic Peer-to-Peer (P2P) Networks. Unlike classic overlays exchanging static, hash-indexed content (e.g., BitTorrent), agentic P2P exchanges dynamic, heterogeneous capabilities that are potentially unsafe, requiring semantic discovery, drift management, and risk-aware execution. We outline the networking foundations for such collaboration. To organize this design space, we propose a plane-based reference architecture decoupling connectivity, semantic discovery, and execution. To support discovery under drift, we introduce signed, soft-state capability descriptors to support intent-aware routing. Finally, to cope with adversarial settings, we present a tiered verification spectrum - ranging from reputation signals (Tier 1) and canary probes (Tier 2) to cryptographic evidence (Tier 3) - that aligns trust costs with task risks. Results from a proxy-level discrete-event simulation suggest improved workflow success under poisoning and drift with modest control-plane overhead.
| Comments: | Accepted for publication in IEEE Communications Magazine |
| Subjects: | Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2603.03753 [cs.NI] |
| (or arXiv:2603.03753v2 [cs.NI] for this version) | |
| https://doi.org/10.48550/arXiv.2603.03753 arXiv-issued DOI via DataCite |
|
| Journal reference: | T. Wang, L. You, J. Tong, C. Zhao, and S. Zhang, "Agentic Peer-to-Peer Networks: From Content Distribution to Capability and Action Sharing," IEEE Communications Magazine, Early Access, pp. 1-7, Oct. 2026 |
| Related DOI: | https://doi.org/10.1109/MCOM.001.2600238
DOI(s) linking to related resources |
Submission history
From: Wang Taotao [view email]
[v1]
Wed, 4 Mar 2026 05:58:44 UTC (6,092 KB)
[v2]
Fri, 9 Oct 2026 12:29:50 UTC (6,010 KB)
来源：arXiv：cs.AI · arxiv.org