Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Open-ended text generation on Law-MT Out of Domain (test)
Loading...
32.17
MAUVE
FoSS
18.1612
21.7981
25.435
29.0719
Feb 11, 2026
MAUVE
Updated 4d ago
Evaluation Results
Method
Method
Links
MAUVE
FoSS
Decoding Strategy=Nucleus
2026.02
32.17
GFlowNets-FT
Decoding Strategy=Nucleus
2026.02
28.62
CoG
Decoding Strategy=Nucleus
2026.02
28.14
FoSS
Decoding Strategy=Greedy
2026.02
27.84
Transformer w/ FT
Decoding Strategy=Nucleus
2026.02
26.85
GFlowNets-FT
Decoding Strategy=Greedy
2026.02
26.49
GDV
Decoding Strategy=Greedy
2026.02
26.35
Transformer w/o FT
Decoding Strategy=Nucleus
2026.02
25.21
GDV
Decoding Strategy=Nucleus
2026.02
24.8
kNN-LM
Decoding Strategy=Nucleus
2026.02
24.75
kNN-LM
Decoding Strategy=Greedy
2026.02
23.31
Transformer w/ FT
Decoding Strategy=Greedy
2026.02
23
CoG
Decoding Strategy=Greedy
2026.02
21.31
RETRO
Decoding Strategy=Nucleus
2026.02
20.35
Transformer w/o FT
Decoding Strategy=Greedy
2026.02
20.32
RETRO
Decoding Strategy=Greedy
2026.02
18.7
Feedback
Search any
task
Search any
task