Transfer Learning Across Languages With Poetry
Transfer learning has been shown to help achieve state of the art results in various natural language processing tasks such as, next sentence prediction and question answering and machine translation. The GPT-2 model has been shown to achieve state of the art results in English language generation; however we propose that it can also be used for various other languages such as Spanish, Ukrainian, Hindi, Bengali, and Assamese through transfer learning. All of these languages belong to the Indo-European language group although geographically dispersed, and therefore share some similarities ...
(For more, see "View full record.")