Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Content Generation on CoGenesis human assessment and automated evaluation
Loading...
50
Tie Rate
SLM finetuned
8.4
19.2
30
40.8
Mar 5, 2024
Tie Rate
Win Rate
Loss Rate
BLEU
ROUGE-L
Updated 4d ago
Evaluation Results
Method
Method
Links
Tie Rate
Win Rate
Loss Rate
BLEU
ROUGE-L
SLM finetuned
Model and Setting=SLM...
2024.03
50
-
-
2.07
13.95
Logits-based CoGen
Model and Setting=Logi...
2024.03
15
32
53
2.3
14.18
Sketch-based CoGen
Model and Setting=Sket...
2024.03
13
27
60
1.81
12.98
LLM w/ context
Model and Setting=LLM...
2024.03
12
38
50
2.61
14.66
LLM w/o context
Model and Setting=LLM...
2024.03
10
3
87
1.51
13.54
Feedback
Search any
task
Search any
task