Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Chit-chat Dialogue Generation on Diamante (test)
Loading...
98.3
Fluency
Top-k
78.228
83.439
88.65
93.861
Jun 12, 2024
Fluency
Relevance
Kappa
Updated 4d ago
Evaluation Results
Method
Method
Links
Fluency
Relevance
Kappa
Top-k
Decoding Setting=DDS
2024.06
98.3
70
0.439
Top-k
Decoding Setting=fixed T
2024.06
97.6
59
0.618
Top-p
Decoding Setting=fixed T
2024.06
92
62.3
0.734
Top-p
Decoding Setting=DDS
2024.06
90.3
60.3
0.655
Typical
Decoding Setting=DDS
2024.06
87.3
54.7
0.431
Typical
Decoding Setting=fixed T
2024.06
84
54.7
0.621
Temperature
Decoding Setting=fixed T
2024.06
80.3
52.7
0.496
Temperature
Decoding Setting=DDS
2024.06
79
50
0.512
Feedback
Search any
task
Search any
task