mGPT: Few-Shot Learners Go Multilingual
Open paper
Benchmark
Model
Rank
Results
cross-lingual-natural-language-inference-on-xnli
mGPT
#5
Accuracy: 40.6
cross-lingual-transfer-on-xcopa
mGPT
#6
Accuracy: 55.5
Rank counts only results with a code link.