Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Data-to-Text Generation on FewShotWeather 1-shot-250 (Seen)
Loading...
2.66
Grammaticality
BLEURT-based self-training
2.5872
2.6061
2.625
2.6439
Oct 16, 2021
Grammaticality
Naturalness
Informativeness
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Grammaticality
Naturalness
Informativeness
Accuracy
BLEURT-based self-training
Training Split=1-shot-...
2021.10
2.66
2.63
80
93
Full data fine-tuning
Training Split=Full da...
2021.10
2.66
2.61
80
95
T5-small Baseline
Training Split=1-shot-...
2021.10
2.59
2.55
81
94
Feedback
Search any
task
Search any
task