Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Factuality Evaluation on FactScore (unlabeled)
Loading...
76.4
US (%)
PaCE
7.656
25.503
43.35
61.197
Jun 6, 2024
US (%)
UAF
Updated 3d ago
Evaluation Results
Method
Method
Links
US (%)
UAF
PaCE
Target Model=LLaMA2-13...
2024.06
76.4
55.1
Prompting
Target Model=LLaMA2-13...
2024.06
47.5
20
OrthoProj
Target Model=LLaMA2-13...
2024.06
43.2
51.7
Vanilla
Target Model=LLaMA2-13...
2024.06
41.8
38.5
PaCE
Target Model=LLaMA2-7B...
2024.06
30.8
73.3
OrthoProj
Target Model=LLaMA2-7B...
2024.06
26.2
45.9
Prompting
Target Model=LLaMA2-7B...
2024.06
20.4
49
Vanilla
Target Model=LLaMA2-7B...
2024.06
15.4
37.4
VecAdd
Target Model=LLaMA2-13...
2024.06
14.9
68.9
VecAdd
Target Model=LLaMA2-7B...
2024.06
10.3
52.2
Feedback
Search any
task
Search any
task