LLM-based Multi-Agent Systems: Techniques and Business Perspectives

TL;DR

The paper examines LaMAS as a practical path to artificial collective intelligence, detailing the technical architecture, interaction protocols, agent training methods, and security considerations that enable autonomous, tool-enabled collaboration among specialized agents. It proposes a multi-layer protocol framework (instruction processing, message exchange, consensus, credit allocation, experience management) and analyzes privacy and monetization dimensions—traffic and intelligence—through business-oriented models and case studies. The work highlights centralized and decentralized architectures, privacy-preserving technologies, and monetization strategies (AaaS, agent marketplaces, Shapley-based revenue sharing) that together support scalable, secure, and monetizable LaMAS ecosystems. Collectively, the findings point to a practical blueprint for deploying LaMAS in real-world settings, balancing privacy, collaboration efficiency, and commercial incentives to achieve artificial collective intelligence in the near future.

Abstract

In the era of (multi-modal) large language models, most operational processes can be reformulated and reproduced using LLM agents. The LLM agents can perceive, control, and get feedback from the environment so as to accomplish the given tasks in an autonomous manner. Besides the environment-interaction property, the LLM agents can call various external tools to ease the task completion process. The tools can be regarded as a predefined operational process with private or real-time knowledge that does not exist in the parameters of LLMs. As a natural trend of development, the tools for calling are becoming autonomous agents, thus the full intelligent system turns out to be a LLM-based Multi-Agent System (LaMAS). Compared to the previous single-LLM-agent system, LaMAS has the advantages of i) dynamic task decomposition and organic specialization, ii) higher flexibility for system changing, iii) proprietary data preserving for each participating entity, and iv) feasibility of monetization for each entity. This paper discusses the technical and business landscapes of LaMAS. To support the ecosystem of LaMAS, we provide a preliminary version of such LaMAS protocol considering technical requirements, data privacy, and business incentives. As such, LaMAS would be a practical solution to achieve artificial collective intelligence in the near future.

Figures (9)

• Figure 1: Illustration of LaMAS.
• Figure 2: Protocol Hierarchy.
• Figure 3: Traffic Monetization.
• Figure 4: Architectures of LaMAS.
• Figure 5: Centralized Architecture of LaMAS.