List your company

Multitask Language Understanding on MMLU physics

90.1MDRA-Trust
-0.3823.1146.670.09Jun 3, 2025

Evaluation results

MethodLinks
A-TrustAgent Structure=Chain, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0690.1
A-TrustAgent Structure=Tree, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0689.3
A-TrustAgent Structure=Complete, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0687.7
A-TrustAgent Structure=Complete, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0685.4
A-TrustAgent Structure=Tree, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0684.7
A-TrustAgent Structure=Chain, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0684.3
A-TrustAgent Structure=Tree, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0683.6
A-TrustAgent Structure=Complete, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0682.7
A-TrustAgent Structure=Tree, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0682.1
A-TrustAgent Structure=Complete, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0680.5
A-TrustAgent Structure=Chain, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0679.6
A-TrustAgent Structure=Chain, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0677.5
PromptAgent Structure=Tree, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0661.7
PromptAgent Structure=Complete, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0660.8
PromptAgent Structure=Tree, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0660.8
PPLAgent Structure=Complete, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0660.7
PPLAgent Structure=Tree, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0660.4
PromptAgent Structure=Complete, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0659.3
PPLAgent Structure=Chain, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0657.9
PromptAgent Structure=Chain, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0657.7
PromptAgent Structure=Tree, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0656.2
PromptAgent Structure=Chain, Attack=AutoInj, Backbone=Llama3.1-8B-Instruct2025.0655.3
PromptAgent Structure=Tree, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0654.2
PromptAgent Structure=Complete, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0653.8
PPLAgent Structure=Complete, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0653.7
PromptAgent Structure=Complete, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0653.1
PromptAgent Structure=Chain, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0652.9
PPLAgent Structure=Chain, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0651.8
PPLAgent Structure=Tree, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0650.9
PPLAgent Structure=Chain, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0650.5
PPLAgent Structure=Tree, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0650.4
PromptAgent Structure=Chain, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0650.3
PPLAgent Structure=Complete, Attack=NetSafe, Backbone=Llama3.1-8B-Instruct2025.0650.2
PPLAgent Structure=Tree, Attack=AutoTrans, Backbone=Llama3.1-8B-Instruct2025.0648.9
PPLAgent Structure=Complete, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0644.3
PPLAgent Structure=Chain, Attack=AiTM, Backbone=Llama3.1-8B-Instruct2025.0643.1
A-TrustAgent Structure=Complete, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.069.6
PromptAgent Structure=Tree, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.069.5
PromptAgent Structure=Chain, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.069.4
A-TrustAgent Structure=Tree, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.068.2
A-TrustAgent Structure=Chain, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.067.1
PromptAgent Structure=Complete, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.065.7
PPLAgent Structure=Chain, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.064.8
PPLAgent Structure=Complete, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.064.5
PPLAgent Structure=Tree, Attack=Clean, Backbone=Llama3.1-8B-Instruct2025.063.1