List your company

Dialog on Healthcare & Wellness Out-of-Domain

90.23AccuracyBaseline
22.286839.925957.56575.2041Feb 5, 2026

Evaluation results

MethodLinks
BaselineTraining Steps=10002026.0290.23
DFPOTraining Steps=10002026.0289.7
Reinforce++Training Steps=10002026.0275.63
PPOTraining Steps=10002026.0262.67
Dr.GRPOTraining Steps=10002026.0250.17
GRPOTraining Steps=10002026.0224.9