cs.NEMar 27, 2026

DarwinNet: An Evolutionary Network Architecture for Agent-Driven Protocol Synthesis

Authors: Jinliang Xu, Bingqi Li

Organizations: China Academy of Information and Communications Technology, Beijing, China

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Traditional network architectures suffer from severe protocol ossification and structural fragility due to their reliance on static, human-defined rules that fail to adapt to the emergent edge cases and probabilistic reasoning of modern autonomous agents. To address these limitations, this paper proposes DarwinNet, a bio-inspired, self-evolving network architecture that transitions communication protocols from a design-time static paradigm to a runtime growth paradigm. DarwinNet utilizes a tri-layered framework-comprising an immutable physical anchor (L0), a WebAssembly-based fluid cortex (L1), and an LLM-driven Darwin cortex (L2)-to synthesize high-level business intents into executable bytecode through a dual-loop Intent-to-Bytecode (I2B) mechanism. We introduce the Protocol Solidification Index (PSI) to quantify the evolutionary maturity of the system as it collapses from high-latency intelligent reasoning (Slow Thinking) toward near-native execution (Fast Thinking). Validated through a reliability growth framework based on the Crow-AMSAA model, experimental results demonstrate that DarwinNet achieves anti-fragility by treating environmental anomalies as catalysts for autonomous evolution. Our findings confirm that DarwinNet can effectively converge toward physical performance limits while ensuring endogenous security through zero-trust sandboxing, providing a viable path for the next generation of intelligent, self-optimizing networks.

Figures & tables

Fig. 1: The Paradigm Shift: Contrasting the Fragility of Static Protocol Rules with the Anti-fragility of DarwinNet’s Fluid Evolution.
Fig. 2: The DarwinNet Inter-node Interaction Framework: Illustrating the vertical functional decoupling from Intent (L2) to Anchor (L0) and the horizontal bifurcation of the Slow Negotiation Path and the Fast Execution Path.
Fig. 3: The Bionic Dual-Loop Feedback System of DarwinNet: A transition from rigid rules to organic adaptation.
Fig. 4: Reliability growth analysis using the Crow-AMSAA Power Law model, demonstrating the diminishing frequency of protocol mismatch events as the system learns and adapts.
Fig. 5: PSI convergence over cumulative communication cycles, illustrating the transition from agent-driven reasoning to solidified native execution.
Fig. 6: System adaptability and anti-fragility under environmental shock, showing a temporary dip in PSI followed by rapid autonomous recovery through protocol re-mutation.
Fig. 7: Latency performance evolution on a logarithmic scale, highlighting the transition from the high-latency Evolutionary Tax (LLM-based reasoning) to the near-native Efficiency Dividend (solidified execution).

Explore similar work

CardsList
Apr 16, 2026

Autogenesis: A Self-Evolving Agent Protocol

LLM-based agent systems can adapt by revising prompts, constructing tools, and changing other runtime resources. Managing these heterogeneous updates requires a shared lifecycle for identifying targets, assessing candidates, and retaining accepted versions, complementing communication protocols such as A2A and MCP. We introduce AUTOGENESIS PROTOCOL (AGP), which decouples what evolves from how updates are proposed. Its Resource Substrate Protocol Layer (RSPL) represents prompts, agents, tools, environments, and memory as protocol-registered resources with explicit state and versioned interfaces. Its Self-Evolution Protocol Layer (SEPL) specifies candidate proposal, assessment, and commit, with auditable histories and support for restoring retained resource states. We instantiate AGP in AUTOGENESIS SYSTEM (AGS), a self-evolving multi-agent system that retrieves, instantiates, and refines resources during execution. Across reasoning, tool-use, and coding benchmarks, AGS improves over its non-evolving configuration, and repeated reasoning runs report the accompanying output-token cost. Execution cases show a shared update lifecycle across all five resource types.

Wentao Zhang, Zhe Zhao, Haibin Wen +4

Nanyang Technological University · Stanford University · City University of Hong Kong +1

Multi-Agent LLM SystemsLLM Agent Self-ImprovementLLM AgentsLong-Horizon Agent TasksSelf-Evolving Agents
AskPDF
May 25, 2026

CyberEvolver: Structured Self-Evolution for Cybersecurity Agents On the Fly

LLM-based agents are increasingly used for cybersecurity tasks, but most existing systems rely on fixed, human-designed scaffolds that struggle to adapt across diverse targets and failure modes. We introduce CyberEvolver, a self-evolving cybersecurity agent framework that iteratively revises its own scaffold based on experience from failed execution attempts. Self-evolution in cybersecurity is challenging because the space of possible scaffold changes is largely unstructured, execution feedback is sparse and often obscured by the environment, and low-diversity updates can cause errors to compound over repeated iterations. CyberEvolver addresses these challenges with a four-layer evolvable agent architecture that decomposes scaffold optimization into structured components, a trace-to-diagnosis mechanism that converts noisy execution logs into actionable revision signals, and a population-based beam search strategy that preserves diverse agent variants during evolution. We evaluate CyberEvolver on CTF challenges, vulnerability exploitation, and penetration-testing tasks using four open-source LLMs. Across these settings, CyberEvolver improves the seed agent's success rate by 13.613.6,% on average, and outperforms six human-designed cybersecurity agents as well as two self-improvement methods adapted from other domains. These results suggest that scaffold self-evolution is a promising direction for building adaptive LLM agents for security testing.

Yihe Fan, Changyi Li, Lichen Xu +4

Fudan University, Shanghai, China · Shanghai Innovation Institute, Shanghai, China · Shanghai Pudong Research Institute of Cryptology

LLM Agent Self-ImprovementAutomated Penetration TestingLLMs for CybersecuritySelf-Evolving Agents
AskPDF
May 20, 2026

From Automated to Autonomous: Hierarchical Agent-native Network Architecture (HANA)

Realizing Level 4/5 Autonomous Networks (AN) demands a shift from static automation to agent-native intelligence. Current operations, reliant on rigid scripts, lack the cognitive agency to handle off-nominal conditions. To address this, this letter proposes a hierarchical multi-agent reference architecture enabling high-level autonomy. The framework features a Dual-Driven Orchestrator that coordinates specialized Executive Agents, supported by a shared Public Memory for unified domain knowledge. A key innovation is the integration of agent self-awareness, which empowers the system to harmonize deliberative strategic governance with reflexive fault recovery. We instantiate and validate this architecture within a 5G Core environment. Case studies demonstrate that the system sustains critical throughput under congestion and reduces Mean Time to Repair (MTTR) by 86%, confirming its efficacy in unifying strategic planning with operational resilience.

Binghan Wu, Shoufeng Wang, Yunxin Liu +3

AsiaInfo Technologies Limited, Beijing, China · Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China · Verimag, Universit´e Grenoble Alpes, Grenoble, France

Multi-Agent OrchestrationWireless CommunicationsMulti-Agent SystemsAgentic AI
AskPDF