LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation

2025-10-06

AI Summary

Current multi-agent large language model (LLM) systems lack reusable and composable memory mechanisms for workflow automation. Method: We propose LEGOMem, a modular procedural memory framework that decomposes historical task trajectories into fine-grained, semantically grounded memory units, dynamically allocating and retrieving them based on agent roles—orchestrator memory enhances task decomposition, while agent-level memory improves execution fidelity and tool-calling accuracy. Contribution/Results: Evaluated on the OfficeBench benchmark, LEGOMem enables small-to-medium-scale LLM teams to significantly narrow the performance gap with stronger models, achieving an average 23.6% improvement in task success rate. This work is the first to systematically characterize the topological design space of memory in multi-agent systems and to demonstrate its critical role in orchestrating planning-execution synergy.

Technology Category

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

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

We introduce LEGOMem, a modular procedural memory framework for multi-agent large language model (LLM) systems in workflow automation. LEGOMem decomposes past task trajectories into reusable memory units and flexibly allocates them across orchestrators and task agents to support planning and execution. To explore the design space of memory in multi-agent systems, we use LEGOMem as a lens and conduct a systematic study of procedural memory in multi-agent systems, examining where memory should be placed, how it should be retrieved, and which agents benefit most. Experiments on the OfficeBench benchmark show that orchestrator memory is critical for effective task decomposition and delegation, while fine-grained agent memory improves execution accuracy. We find that even teams composed of smaller language models can benefit substantially from procedural memory, narrowing the performance gap with stronger agents by leveraging prior execution traces for more accurate planning and tool use. These results position LEGOMem as both a practical framework for memory-augmented agent systems and a research tool for understanding memory design in multi-agent workflow automation.

Problem

• Developing modular procedural memory for multi-agent LLM workflow systems
• Exploring memory placement and retrieval strategies in multi-agent systems
• Enhancing planning and execution accuracy through reusable memory units

Innovation

• Modular memory decomposes task trajectories into reusable units
• Flexible memory allocation supports multi-agent planning and execution
• Procedural memory narrows performance gap between smaller and larger models