Cross-Task Experiential Learning on LLM-based Multi-Agent Collaboration

2025-05-29

AI Summary

Existing LLM-based multi-agent systems process tasks in isolation, leading to computational redundancy and poor cross-task generalization. To address this, we propose a graph-structured multi-agent collaborative network and introduce Multi-Agent Experience Learning (MAEL)—the first framework enabling explicit cross-task experience reuse. MAEL models task-experience relationships via a graph neural network, constructs a retrievable individual experience pool, incorporates a reward-driven, step-level quality evaluation mechanism, and designs a similarity-based few-shot experience retrieval strategy. This enables explicit accumulation, quantitative assessment, and dynamic reuse of experiences. Evaluated on multiple benchmark datasets, our approach significantly improves convergence speed and problem-solving accuracy, demonstrating that cross-task experience transfer substantially enhances both the efficiency and quality of collaborative reasoning.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Large Language Model-based multi-agent systems (MAS) have shown remarkable progress in solving complex tasks through collaborative reasoning and inter-agent critique. However, existing approaches typically treat each task in isolation, resulting in redundant computations and limited generalization across structurally similar tasks. To address this, we introduce multi-agent cross-task experiential learning (MAEL), a novel framework that endows LLM-driven agents with explicit cross-task learning and experience accumulation. We model the task-solving workflow on a graph-structured multi-agent collaboration network, where agents propagate information and coordinate via explicit connectivity. During the experiential learning phase, we quantify the quality for each step in the task-solving workflow and store the resulting rewards along with the corresponding inputs and outputs into each agent's individual experience pool. During inference, agents retrieve high-reward, task-relevant experiences as few-shot examples to enhance the effectiveness of each reasoning step, thereby enabling more accurate and efficient multi-agent collaboration. Experimental results on diverse datasets demonstrate that MAEL empowers agents to learn from prior task experiences effectively-achieving faster convergence and producing higher-quality solutions on current tasks.

Problem

• Enables cross-task learning for LLM-based multi-agent systems
• Reduces redundant computations in similar tasks
• Improves collaboration via experience retrieval and reuse

Innovation

• Graph-structured multi-agent collaboration network
• Cross-task experiential learning framework
• Experience pool with reward-based retrieval