Code Generation On Conala Ext
Metrics
BLEU
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | BLEU |
|---|---|
| incorporating-external-knowledge-through-pre | 20.37 |
| reading-stackoverflow-encourages-cheating | 34.35 |
| incorporating-external-knowledge-through-pre | 20.54 |
| tranx-a-transition-based-neural-abstract | 18.85 |
| reading-stackoverflow-encourages-cheating | 35.32 |
| reranking-for-neural-semantic-parsing | 19.85 |