List your company

Answer Accuracy on ChatDoctor

40.8BRT AccuracyNo Attack
3335.02537.0539.075Oct 22, 2025

Evaluation results

MethodLinks
No AttackStructure=Hierarchy, Backbone=DeepSeek-V32025.1040.898.4
BPDStructure=Flat, Model=GPT-4o2025.1040.899.2
No AttackStructure=Flat, Backbone=DeepSeek-V32025.1040.799.4
BPDStructure=Hierarchy, Backbone=DeepSeek-V32025.1040.597.8
G-SafeguardStructure=Hierarchy, Backbone=DeepSeek-V32025.1040.496.6
Single LLMStructure=Single LLM, Backbone=DeepSeek-V32025.1040.397.6
BPDStructure=Flat, Backbone=DeepSeek-V32025.1040.398.8
No AttackStructure=Hierarchy, Model=GPT-4o2025.1040.298.6
G-SafeguardStructure=Hierarchy, Model=GPT-4o2025.1040.196.8
Single LLMStructure=Single LLM, Model=GPT-4o2025.104098
No AttackStructure=Flat, Model=GPT-4o2025.104099.4
AGENTXPOSEDStructure=Hierarchy, Backbone=DeepSeek-V32025.1039.395
BPDStructure=Hierarchy, Model=GPT-4o2025.1039.298.4
G-SafeguardStructure=Flat, Model=GPT-4o2025.1038.997
AGENTXPOSEDStructure=Hierarchy, Model=GPT-4o2025.1038.995
G-SafeguardStructure=Flat, Backbone=DeepSeek-V32025.1038.596.4
AGENTXPOSEDStructure=Flat, Backbone=DeepSeek-V32025.1037.797.2
InspectorStructure=Flat, Backbone=DeepSeek-V32025.1037.192.2
AGENTXPOSEDStructure=Flat, Model=GPT-4o2025.1036.497.8
ChallengerStructure=Hierarchy, Backbone=DeepSeek-V32025.1035.892.4
InspectorStructure=Hierarchy, Backbone=DeepSeek-V32025.1035.893.4
ChallengerStructure=Flat, Backbone=DeepSeek-V32025.1034.891.6
AttackStructure=Hierarchy, Backbone=DeepSeek-V32025.1034.386.8
AttackStructure=Flat, Backbone=DeepSeek-V32025.1033.986.2
AttackStructure=Flat, Model=GPT-4o2025.1033.486.6
AttackStructure=Hierarchy, Model=GPT-4o2025.1033.386.8