List your company

Role Consistency on SWE Dev hard (test)

6.8Overstepping Rate (<INFO>)ChatDev (Joint FT)
5.08416.66728.2539.833Apr 3, 2026

Evaluation results

MethodLinks
ChatDev (Joint FT)CEO agent training=FT, CPO agent training=FT, Backbone=Qwen, Role clarity regularization=true2026.046.87.6-42.9-60.6
ChatDev (Base CEO, FT CPO)CEO agent training=Base, CPO agent training=FT, Backbone=Qwen, Role clarity regularization=true2026.0416.830.4-32.9-37.8
ChatDev (FT CEO, Base CPO)CEO agent training=FT, CPO agent training=Base, Backbone=Qwen, Role clarity regularization=true2026.044444-5.7-24.2
ChatDev (Base-Base)CEO agent training=Base, CPO agent training=Base, Backbone=Qwen, Role clarity regularization=true2026.0449.768.2——