List your company

Dialog on Financial Services Out-of-Domain

84.3AccuracyDFPO
27.172842.003956.83571.6661Feb 5, 2026

Evaluation results

MethodLinks
DFPOTraining Steps=10002026.0284.3
Reinforce++Training Steps=10002026.0283.53
BaselineTraining Steps=10002026.0282.7
PPOTraining Steps=10002026.0261.47
Dr.GRPOTraining Steps=10002026.0246.6
GRPOTraining Steps=10002026.0229.37