AWorld: Dynamic Multi-Agent System with Stable Maneuvering for Robust GAIA Problem Solving

2025-08-13

AI Summary

To address the declining reliability and accuracy of intelligent agents in multi-tool invocation scenarios—caused by context expansion and noisy outputs—this paper proposes AWorld, a dynamic multi-agent framework. Methodologically, AWorld introduces a novel “executor agent–guardian agent” collaborative architecture, integrating dynamic supervision and real-time regulation mechanisms to enable autonomous stepwise correction and context purification. Built upon large language models, the framework tightly couples tool invocation, adaptive context management, and on-the-fly intervention, thereby substantially enhancing reasoning robustness in complex tasks. Empirically, AWorld achieves state-of-the-art performance on the GAIA benchmark, becoming the first open-source system to top its leaderboard; it significantly outperforms both single-agent baselines and standard tool-augmented approaches in both accuracy and system stability.

Technology Category

• Multiagent Systems: Adversarial Agents
• Intelligent Robots: Multi-Robot Systems
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

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

Abstract

The rapid advancement of large language models (LLMs) has empowered intelligent agents to leverage diverse external tools for solving complex real-world problems. However, as agents increasingly depend on multiple tools, they encounter new challenges: extended contexts from disparate sources and noisy or irrelevant tool outputs can undermine system reliability and accuracy. These challenges underscore the necessity for enhanced stability in agent-based systems. To address this, we introduce dynamic supervision and maneuvering mechanisms, constructing a robust and dynamic Multi-Agent System (MAS) architecture within the AWorld framework. In our approach, the Execution Agent invokes the Guard Agent at critical steps to verify and correct the reasoning process, effectively reducing errors arising from noise and bolstering problem-solving robustness. Extensive experiments on the GAIA test dataset reveal that our dynamic maneuvering mechanism significantly improves both the effectiveness and stability of solutions, outperforming single-agent system (SAS) and standard tool-augmented systems. As a result, our dynamic MAS system achieved first place among open-source projects on the prestigious GAIA leaderboard. These findings highlight the practical value of collaborative agent roles in developing more reliable and trustworthy intelligent systems.

Problem

• Enhances stability in multi-agent systems handling diverse tools
• Reduces errors from noisy tool outputs in reasoning processes
• Improves robustness and accuracy in GAIA problem-solving tasks

Innovation

• Dynamic supervision and maneuvering mechanisms
• Execution Agent invokes Guard Agent
• Multi-Agent System enhances stability