OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation

Large Language Model (LLM)-based multi-agent systems show promise for automating real-world tasks but struggle to transfer across domains due to their domain-specific nature. Current approaches face two critical shortcomings: they require complete architectural redesign and full retraining of all components when applied to new domains. We introduceWorkforce, a hierarchical multi-agent framework that decouples strategic planning from specialized execution through a modular architecture comprising:(i)adomain-agnosticPlannerfor task decomposition,(ii)aCoordinatorfor subtask management, and(iii)specializedWorkerswithdomain-specifictool-calling capabilities. This decoupling enables cross-domain transferability during both inference and training phases: During inference, Workforce seamlessly adapts to new domains by adding or modifying worker agents; For training, we introduceOptimized Workforce Learning (OWL), which improves generalization across domains by optimizing a domain-agnostic planner with reinforcement learning from real-world feedback. To validate our approach, we evaluate Workforce on the GAIA benchmark, covering various realistic, multi-domain agentic tasks. Experimental results demonstrate Workforce achieves open-source state-of-the-art performance (69.70%), outperforming commercial systems like OpenAI's Deep Research by2.34%. More notably, our OWL-trained 32B model achieves52.73%accuracy (+16.37%) and demonstrates performance comparable to GPT-4o on challenging tasks. To summarize, by enabling scalable generalization and modular domain transfer, our work establishes a foundation for the next generation of general-purpose AI assistants.Our code is available atAnonymous URL, and our data is available atAnonymous URL.