BLOOMZ

Reported on 5 benchmarks across 4 tasks · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing5 results

Question AnsweringonStoryCloze
Accuracy· 2022-11-03
96.3
SOTA
Crosslingual Generalization through Multitask Finetuning arXiv:2211.01786
Cross-LingualonXCOPA
Accuracy· 2022-11-03
75.5
best: 94.4 (PaLM 2 (few-shot))
Crosslingual Generalization through Multitask Finetuning arXiv:2211.01786
Coreference ResolutiononXWinograd EN
Accuracy· 2022-11-03
69.08
best: 81.29 (mT0-13B)
Crosslingual Generalization through Multitask Finetuning arXiv:2211.01786
Coreference ResolutiononXWinograd FR
Accuracy· 2022-11-03
68.67
best: 78.31 (mT0-13B)
Crosslingual Generalization through Multitask Finetuning arXiv:2211.01786
Cross-Lingual TransferonXCOPA
Accuracy· 2022-11-03
75.5
best: 94.4 (PaLM 2 (few-shot))
Crosslingual Generalization through Multitask Finetuning arXiv:2211.01786