Towards a Science of Scaling Agent Systems
BeginnerYubin Kim, Ken Gu et al.Dec 9arXiv
Multi-agent AI teams are not automatically better; their success depends on matching the teamβs coordination style to the jobβs structure.
#multi-agent systems#agentic evaluation#scaling laws