Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MiTa: A Hierarchical Multi-Agent Collaboration Framework with Memory-integrated and Task Allocation

XiaoJie Zhang, JianHan Wu, Xiaoyang Qu, Jianzong Wang

TL;DR

By combining task allocation with episodic memory, MiTa attains a clearer understanding of the task and facilitates globally consistent task distribution, and achieves superior efficiency and adaptability in complex multi-agent cooperation over strong baseline methods.

Abstract

Recent advances in large language models (LLMs) have substantially accelerated the development of embodied agents. LLM-based multi-agent systems mitigate the inefficiency of single agents in complex tasks. However, they still suffer from issues such as memory inconsistency and agent behavioral conflicts. To address these challenges, we propose MiTa, a hierarchical memory-integrated task allocative framework to enhance collaborative efficiency. MiTa organizes agents into a manager-member hierarchy, where the manager incorporates additional allocation and summary modules that enable (1) global task allocation and (2) episodic memory integration. The allocation module enables the manager to allocate tasks from a global perspective, thereby avoiding potential inter-agent conflicts. The summary module, triggered by task progress updates, performs episodic memory integration by condensing recent collaboration history into a concise summary that preserves long-horizon context. By combining task allocation with episodic memory, MiTa attains a clearer understanding of the task and facilitates globally consistent task distribution. Experimental results confirm that MiTa achieves superior efficiency and adaptability in complex multi-agent cooperation over strong baseline methods.

MiTa: A Hierarchical Multi-Agent Collaboration Framework with Memory-integrated and Task Allocation

TL;DR

By combining task allocation with episodic memory, MiTa attains a clearer understanding of the task and facilitates globally consistent task distribution, and achieves superior efficiency and adaptability in complex multi-agent cooperation over strong baseline methods.

Abstract

Recent advances in large language models (LLMs) have substantially accelerated the development of embodied agents. LLM-based multi-agent systems mitigate the inefficiency of single agents in complex tasks. However, they still suffer from issues such as memory inconsistency and agent behavioral conflicts. To address these challenges, we propose MiTa, a hierarchical memory-integrated task allocative framework to enhance collaborative efficiency. MiTa organizes agents into a manager-member hierarchy, where the manager incorporates additional allocation and summary modules that enable (1) global task allocation and (2) episodic memory integration. The allocation module enables the manager to allocate tasks from a global perspective, thereby avoiding potential inter-agent conflicts. The summary module, triggered by task progress updates, performs episodic memory integration by condensing recent collaboration history into a concise summary that preserves long-horizon context. By combining task allocation with episodic memory, MiTa attains a clearer understanding of the task and facilitates globally consistent task distribution. Experimental results confirm that MiTa achieves superior efficiency and adaptability in complex multi-agent cooperation over strong baseline methods.
Paper Structure (11 sections, 4 equations, 3 figures, 2 tables, 1 algorithm)

This paper contains 11 sections, 4 equations, 3 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Task Formulation
  3. method
  4. Overall Framework
  5. Negotiation-Aware Centralized Task Allocation
  6. LLM-Driven Episodic Memory Integration
  7. Experiment
  8. Experiment Setup
  9. Main Results and Analysis
  10. Ablation Study
  11. CONCLUSIONS

Figures (3)

  • Figure 1: Overview of the MiTa framework. MiTa organizes agents into two roles: manager and member. (a) The overall workflow of MiTa, including the structures of both managers and members. (b) Detailed workflow of the manager agent. (c) Details of the Summary modules. (d) Structure of the prompt used in the allocation module.
  • Figure 2: Example of generating a collaborative summary
  • Figure 3: (a) MiTa performance under symbolic observation across different LLMs. (b) Robustness under symbolic observation (3 agents) in computationally limited settings.