Marvelous Agglutinative Language Effect on Cross Lingual Transfer Learning
This finding could change training strategies for cross-lingual transfer learning, addressing the curse of multilinguality in NLP.
The paper tackles the problem of selecting languages for multilingual model training by demonstrating that using agglutinative languages like Korean is more effective for cross-lingual transfer learning than relying on similar language structures.
As for multilingual language models, it is important to select languages for training because of the curse of multilinguality. It is known that using languages with similar language structures is effective for cross lingual transfer learning. However, we demonstrate that using agglutinative languages such as Korean is more effective in cross lingual transfer learning. This is a great discovery that will change the training strategy of cross lingual transfer learning.