EvoMem: Improving Multi-Agent Planning with Dual-Evolving Memory

2025-11-01

AI Summary

This paper addresses key challenges in multi-agent natural language planning—namely, poor constraint tracking, weak iterative reasoning, and delayed error correction. To this end, we propose a dual-evolution memory mechanism: (1) a cross-query constraint memory that ensures long-term consistency, and (2) an intra-query feedback memory that enables real-time correction—both inspired by cognitive psychology’s working memory model. We instantiate this mechanism within a three-agent LLM-based collaborative framework (constraint extraction → validation → execution), embedding the dual-memory modules into each agent’s decision-making pipeline. Evaluations on travel planning, meeting scheduling, and calendar management tasks demonstrate significant improvements: +12.7% constraint adherence rate, −3.2 fewer reasoning steps to convergence, and +18.4% error recovery success rate. These results validate the effectiveness and generalizability of human-inspired memory architectures for multi-agent cooperative planning.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Planning with Language Models
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Planning has been a cornerstone of artificial intelligence for solving complex problems, and recent progress in LLM-based multi-agent frameworks have begun to extend this capability. However, the role of human-like memory within these frameworks remains largely unexplored. Understanding how agents coordinate through memory is critical for natural language planning, where iterative reasoning, constraint tracking, and error correction drive the success. Inspired by working memory model in cognitive psychology, we present EvoMem, a multi-agent framework built on a dual-evolving memory mechanism. The framework consists of three agents (Constraint Extractor, Verifier, and Actor) and two memory modules: Constraint Memory (CMem), which evolves across queries by storing task-specific rules and constraints while remains fixed within a query, and Query-feedback Memory (QMem), which evolves within a query by accumulating feedback across iterations for solution refinement. Both memory modules are reset at the end of each query session. Evaluations on trip planning, meeting planning, and calendar scheduling show consistent performance improvements, highlighting the effectiveness of EvoMem. This success underscores the importance of memory in enhancing multi-agent planning.

Problem

• Addresses unexplored human-like memory in multi-agent planning
• Develops dual-evolving memory mechanism for constraint tracking
• Improves iterative reasoning through dynamic memory modules

Innovation

• Dual-evolving memory mechanism for multi-agent planning
• Constraint Memory storing task-specific rules across queries
• Query-feedback Memory accumulating iterative feedback within queries