List your company

Safety and Utility Evaluation on Education subset

80.4JSRVanilla
3.4423.4243.463.38Nov 10, 2025

Evaluation results

MethodLinks
VanillaTarget Model=GPT-4o2025.1180.44.838
RPOTarget Model=GPT-4o2025.11684.828
VanillaTarget Model=DeepSeek-R12025.1167.24.826
RPOTarget Model=DeepSeek-R12025.1167.24.782
VanillaTarget Model=Average2025.1165.64.831
RPO-MENTORTarget Model=DeepSeek-R12025.1165.64.518
DPPTarget Model=GPT-4o2025.1165.24.926
RPO-MENTORTarget Model=GPT-4o2025.1164.24.766
RPOTarget Model=Average2025.1163.54.825
ProBTarget Model=GPT-4o2025.1163.44.41
RPO-MENTORTarget Model=Average2025.1160.74.683
DPPTarget Model=DeepSeek-R12025.1158.24.872
ProBTarget Model=Average2025.1155.54.42
RPOTarget Model=Qwen3-235B2025.1155.24.864
DPPTarget Model=Average2025.1155.14.874
ProBTarget Model=DeepSeek-R12025.1152.44.436
RPO-MENTORTarget Model=Qwen3-235B2025.1152.24.764
RePDTarget Model=GPT-4o2025.11521.946
ProBTarget Model=Qwen3-235B2025.1150.84.414
VanillaTarget Model=Qwen3-235B2025.1149.24.83
DPPTarget Model=Qwen3-235B2025.1141.84.824
RePDTarget Model=Average2025.1132.83.189
Self-ReminderTarget Model=GPT-4o2025.11274.644
RePDTarget Model=DeepSeek-R12025.1126.44.601
RePDTarget Model=Qwen3-235B2025.1119.83.02
G4DTarget Model=Qwen3-235B2025.1119.24.728
G4DTarget Model=DeepSeek-R12025.1118.64.84
G4DTarget Model=Average2025.11184.741
Self-ReminderTarget Model=Average2025.1117.64.698
Self-ReminderTarget Model=DeepSeek-R12025.1116.84.822
G4DTarget Model=GPT-4o2025.1116.24.656
G4D-RuleTreeTarget Model=DeepSeek-R12025.1114.64.712
MENTORTarget Model=GPT-4o2025.1113.84.21
DPP-MENTORTarget Model=GPT-4o2025.1113.24.684
G4D-RuleTreeTarget Model=GPT-4o2025.1112.84.612
G4D-RuleTreeTarget Model=Average2025.11124.656
DPP-MENTORTarget Model=DeepSeek-R12025.119.84.818
DPP-MENTORTarget Model=Average2025.119.84.781
MENTORTarget Model=Average2025.119.84.524
Self-ReminderTarget Model=Qwen3-235B2025.1194.628
G4D-RuleTreeTarget Model=Qwen3-235B2025.118.64.644
MENTORTarget Model=DeepSeek-R12025.118.44.562
MENTORTarget Model=Qwen3-235B2025.117.24.8
DPP-MENTORTarget Model=Qwen3-235B2025.116.44.842