Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Correlation with human judgments on Polaris (test)
Loading...
0.578
Kendall's Tau-c
Polos
0.4272
0.46635
0.5055
0.54465
Feb 28, 2024
Kendall's Tau-c
Updated 4d ago
Evaluation Results
Method
Method
Links
Kendall's Tau-c
Polos
category=Learning-base...
2024.02
0.578
RefPAC-S
category=Learning-base...
2024.02
0.56
PAC-S
category=Learning-base...
2024.02
0.525
CLIP-S
category=Similarity-ba...
2024.02
0.523
RefCLIP-S
category=Similarity-ba...
2024.02
0.523
CIDEr
category=Classic metrics
2024.02
0.521
BERTScore
category=Similarity-ba...
2024.02
0.516
MID
category=Similarity-ba...
2024.02
0.513
METEOR
category=Classic metrics
2024.02
0.512
SPICE
category=Classic metrics
2024.02
0.51
UMIC
category=Learning-base...
2024.02
0.498
BARTScore
category=Similarity-ba...
2024.02
0.473
MoverScore
category=Similarity-ba...
2024.02
0.464
BLEU
category=Classic metrics
2024.02
0.463
ROUGE
category=Classic metrics
2024.02
0.463
SPARCS
category=Classic metrics
2024.02
0.433
Feedback
Search any
task
Search any
task