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

TL;DR

This paper introduces Workforce, a modular multi-agent framework with a novel training method called OWL, enabling cross-domain transfer and improved performance in real-world task automation.

Contribution

We propose Workforce, a hierarchical, modular multi-agent system with OWL training for enhanced cross-domain transferability and generalization in real-world tasks.

Findings

Achieves state-of-the-art 69.70% performance on GAIA benchmark.

OWL-trained 32B model reaches 52.73% accuracy, close to GPT-4o.

Demonstrates effective cross-domain transfer during inference and training.

Abstract

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 introduce Workforce, a hierarchical multi-agent framework that decouples strategic planning from specialized execution through a modular architecture comprising: (i) a domain-agnostic Planner for task decomposition, (ii) a Coordinator for subtask management, and (iii) specialized Workers with domain-specific tool-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 introduce Optimized 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 by 2.34%. More notably, our OWL-trained 32B model achieves 52.73% 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.