Translation On Iwslt 2017

BLEU

Results

Performance results of various models on this benchmark

Comparison Table

Model Name	BLEU
hierarchical-prompting-taxonomy-a-universal	0.23539