Open access
Jul 2026
Capable language models can outgrow the benefits of collaboration
A controlled study of large language model agents across 260 configurations shows when multi-agent collaboration helps or hurts performance, and introduces a predictive model that selects the best architecture in 87% of held-out within-domain configurations.
Y. Kim, Ken Gu, Chanwoo Park et al.
· Nature Machine Intelligence · 6 citations