Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Constrained Text Generation on CommonGen 500 randomly sampled data (test)
Loading...
32.53
BLEU-4
MORE
19.3532
22.7741
26.195
29.6159
Feb 21, 2024
BLEU-4
CIDEr
SPICE
Updated 4d ago
Evaluation Results
Method
Method
Links
BLEU-4
CIDEr
SPICE
MORE
backbone=OPT-2.7b, ret...
2024.02
32.53
17.3
32.81
MORE
backbone=OPT-2.7b, ret...
2024.02
31.81
17.08
31.81
GPT-4
n-shot=3-shot
2024.02
30
16.41
29.05
GPT-4
n-shot=0-shot
2024.02
28.53
16.52
30.53
GPT-3.5
n-shot=3-shot
2024.02
28.35
16.14
29.13
GPT-4
n-shot=0-shot, length...
2024.02
27.87
16.89
29.11
GPT-3.5
n-shot=0-shot, length...
2024.02
25.54
15.62
26.75
GPT-3.5
n-shot=0-shot
2024.02
22.97
13.93
27.25
GPT-4
n-shot=0-shot, retriev...
2024.02
19.86
12.43
26.2
Feedback
Search any
task
Search any
task