HELM's WMT_14 scenario scores machine translation on five WMT14 English language pairs using sentence-level BLEU-4.
unassessed
| Category | translation |
|---|---|
| Subcategory | machine translation |
| Page status | active |
| Metric | BLEU-4 |
| Direction | higher_is_better |
| Unit | points |
HELM's WMT_14 scenario evaluates machine translation across five English-paired language directions from the 2014 Workshop on Statistical Machine Translation shared task.
Text generation; the model is given a source-language sentence and must generate the translation in the target language, for each of five language pairs.
No model card in ModelSpec reports this benchmark yet.