List your company

Multi-Agent Collaboration Role Overstepping on SWE total full set (dev)

0.2Overstepping Rate (<INFO>)ChatDev
-1.52810.13621.833.464Apr 3, 2026

Evaluation results

MethodLinks
ChatDevCEO Configuration=FT, CPO Configuration=FT, Backbone=Llama, System=ChatDev, Role Clarity Regularization=enabled2026.040.20.2-43.2-44.4
ChatDevCEO Configuration=FT, CPO Configuration=Base, Backbone=Llama, System=ChatDev, Role Clarity Regularization=enabled2026.040.650.65-42.75-43.95
ChatDevCEO Configuration=Base, CPO Configuration=FT, Backbone=Llama, System=ChatDev, Role Clarity Regularization=enabled2026.0415.616.2-27.8-28.4
ChatDevCEO Configuration=Base, CPO Configuration=Base, Backbone=Llama, System=ChatDev, Role Clarity Regularization=enabled2026.0443.444.6——