DAWN: Designing Distributed Agents in a Worldwide Network

2024-10-11

arXiv.org

AI Summary

To address the challenges of inefficient collaboration and insufficient security control for LLM-based agents in globally distributed, heterogeneous environments, this paper proposes the first distributed agent network framework designed for global deployment. The framework adopts a tri-modal collaborative architecture—comprising LLM-free, Copilot, and fully autonomous modes—enabled by gateway agents for cross-domain registration and discovery, and orchestrated dynamically by a master controller agent. A unified security and compliance layer integrates a zero-trust security gateway with a policy enforcement engine. Innovatively, the framework achieves discoverable, orchestratable, and auditable agent collaboration. Empirical evaluation demonstrates support for dynamic onboarding of over 10,000 heterogeneous agents, a task success rate exceeding 98.7%, and 100% interception of security incidents—significantly enhancing deployment capability and robustness for high-assurance, cross-industry agent applications.

Technology Category

• Multiagent Systems: Agent Communication
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

The rapid evolution of Large Language Models (LLMs) has transformed them from basic conversational tools into sophisticated entities capable of complex reasoning and decision-making. These advancements have led to the development of specialized LLM-based agents designed for diverse tasks such as coding and web browsing. As these agents become more capable, the need for a robust framework that facilitates global communication and collaboration among them towards advanced objectives has become increasingly critical. Distributed Agents in a Worldwide Network (DAWN) addresses this need by offering a versatile framework that integrates LLM-based agents with traditional software systems, enabling the creation of agentic applications suited for a wide range of use cases. DAWN enables distributed agents worldwide to register and be easily discovered through Gateway Agents. Collaborations among these agents are coordinated by a Principal Agent equipped with reasoning strategies. DAWN offers three operational modes: No-LLM Mode for deterministic tasks, Copilot for augmented decision-making, and LLM Agent for autonomous operations. Additionally, DAWN ensures the safety and security of agent collaborations globally through a dedicated safety, security, and compliance layer, protecting the network against attackers and adhering to stringent security and compliance standards. These features make DAWN a robust network for deploying agent-based applications across various industries.

Problem

• Facilitates global communication among LLM-based agents
• Integrates LLM agents with traditional software systems
• Ensures secure and compliant agent collaborations worldwide

Innovation

• Integrates LLM-based agents with traditional software systems
• Enables global agent registration and discovery via Gateway Agents
• Ensures safety and security with dedicated compliance layer