Memory system boosts teamwork in multi agent ai systems
CoMem: Collective-Individual Memory Synergy for Evolutionary Multi-Agent Systems
Artificial Intelligence
Summary
Many AI agents working together need good memory systems to help them learn and cooperate better. The authors found that using a mix of private memories unique to each agent and shared memories common to all helps prevent confusion and forgetting. They built a system called CoMem that keeps individual useful memories, shares only well-tested ideas with the group, and lets agents pick from both kinds using smart methods. Tests showed CoMem helped agents perform better and kept their memories clean.
What this means in practice
- •For multi agent system developers: Build cooperative AI agents that learn efficiently by combining private and shared memories to improve joint task performance.
- •For robotics teams: Enable robot groups to maintain diverse yet shared knowledge for robust collective decision making during complex tasks.
Authors
Chengxin Yu, Zhaoxin Fan, Faguo Wu, Hongwei Zheng, Yun Zhou, Zhiyu Li
Abstract
Designing effective memory mechanisms is crucial for advancing LLM-driven Multi-Agent Systems (MAS), helping agents learn together and perform better over time. While recent work has led to strong cooperation skills, most methods still use flat, unstructured memories, which easily get filled with noise and erase differences between agents. To address this, we introduce the concept of collective-individual memory synergy and propose CoMem, an architecture that unifies both private experience and shared knowledge for multi-agent learning. CoMem features:(i) Private Experience Sedimentation, which lets each agent keep and update its own useful memories over time;(ii) Collective Wisdom Curation, which carefully selects only widely proven ideas to be shared among agents;(iii)Parallel Dual-Stream Retrieval, which allows agents to draw both from their own memory and the group's wisdom, using clustering to ensure diversity.Experiments on ALFWorld and PDDL benchmarks show that CoMem achieves strong overall performance and robustly avoids memory pollution.