Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Knowledge Insertion on WikiData recent (test)
Loading...
100
Edit Success Rate
FT-M
43.4552
58.1351
72.815
87.4949
Jun 17, 2024
Edit Success Rate
Portability
Locality
Fluency
Perplexity (PPL)
Updated 4d ago
Evaluation Results
Method
Method
Links
Edit Success Rate
Portability
Locality
Fluency
Perplexity (PPL)
FT-M
Backbone=Llama2-7b-chat
2024.06
100
59.28
41.54
587.17
70.64
ICE
Backbone=Llama2-7b-chat
2024.06
100
61.02
46.39
585.58
34.08
ROME
Backbone=Llama2-7b-chat
2024.06
97.25
36.58
30.4
581
107.47
MEMIT
Backbone=Llama2-7b-chat
2024.06
97.03
37
29.28
573.06
87.17
FT-L
Backbone=Llama2-7b-chat
2024.06
45.63
34.73
34.8
558.91
68.92
Feedback
Search any
task
Search any
task