Controlling Large Language Model-based Agents for Large-Scale Decision-Making: An Actor-Critic Approach
This addresses coordination and efficiency issues in large-scale decision-making for multi-agent systems, representing an incremental improvement by integrating actor-critic methods with LLMs.
The paper tackles the challenges of hallucination and coordination in large-scale multi-agent systems using LLMs, proposing the LLaMAC framework which shows considerable advantages in system resource allocation and robot grid transportation evaluations.
The remarkable progress in Large Language Models (LLMs) opens up new avenues for addressing planning and decision-making problems in Multi-Agent Systems (MAS). However, as the number of agents increases, the issues of hallucination in LLMs and coordination in MAS have become increasingly prominent. Additionally, the efficient utilization of tokens emerges as a critical consideration when employing LLMs to facilitate the interactions among a substantial number of agents. In this paper, we develop a modular framework called LLaMAC to mitigate these challenges. LLaMAC implements a value distribution encoding similar to that found in the human brain, utilizing internal and external feedback mechanisms to facilitate collaboration and iterative reasoning among its modules. Through evaluations involving system resource allocation and robot grid transportation, we demonstrate the considerable advantages afforded by our proposed approach.