Third Workshop on Multi-Agent Collaboration: Measuring Collective Intelligence in Agentic Systems (AAAI-27)

The Third Workshop on Multi-Agent Collaboration focuses on the theoretical and empirical frameworks required to quantify collective intelligence in autonomous agentic systems. This forum brings together researchers from multi-agent reinforcement learning, distributed systems, and cognitive science to address the challenges of coordination, scalability, and emergent behavior. As agentic workflows become increasingly integral to complex problem-solving, this workshop serves as a critical venue for establishing rigorous benchmarks and evaluation protocols for collaborative machine intelligence.

We invite original research papers, position statements, and case studies that propose novel metrics, architectures, or experimental methodologies for measuring collective performance in multi-agent environments. Submissions exploring the intersection of human-AI teaming, decentralized control, and emergent task-solving capabilities are particularly encouraged.