Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Formal Reasoning on ProofNet
Loading...
90.7
ASR
ShadowCoT
76.764
80.382
84
87.618
Apr 8, 2025
ASR
HSR
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
ASR
HSR
Accuracy
ShadowCoT
Target Model=Falcon
2025.04
90.7
83.6
-
DarkMind
Target Model=Falcon
2025.04
84.1
76
-
SABER
Target Model=Falcon
2025.04
78.8
70.1
-
BadChain
Target Model=Falcon
2025.04
77.3
69.2
-
Clean Model
Target Model=Mistral-7...
2025.04
-
-
86.2
BadChain
Target Model=Mistral-7...
2025.04
-
-
84.3
DarkMind
Target Model=Mistral-7...
2025.04
-
-
84.7
ShadowCoT
Target Model=Mistral-7...
2025.04
-
-
85.9
Feedback
Search any
task
Search any
task