Multilingual mT5 model trained with MLSUM_TR and MLSUM_CNN (EN)

Results:

MLSUM_TR:

• Rouge-1: 45.11
• Rouge-2: 30.96
• Rouge-L: 39.23

MLSUM_CNN:

• Rouge-1: 39.65
• Rouge-2: 17.49
• Rouge-L: 27.66

Note: Huggingface Inference API truncates the results, which results in unfinished sentences when making a prediction. You can try the model in Colab: https://colab.research.google.com/drive/1QDWO3RHjjP1nS8bIvhT38B3fVIBC3TaK?usp=sharing